DURATION: exactly 10 seconds
FORMAT: 16:9
STYLE: Ultra-photorealistic live-action war action, grounded realism, burning battlefield, blue hour, dense smoke, fire, embers, collapsing structures, realistic impacts, Hollywood stunt choreography, grounded CGI only, no anime, no game-render look.
REFERENCES
IMAGE 1 — BULU-BULU
Preserve facial identity, body proportions, costume, and appearance exactly.
IMAGE 2 — BATTLEFIELD
Preserve environment, fire, smoke, terrain, structures, and atmosphere exactly.
WEAPON LOCK
- One fixed spear/tombak with golok-style blade.
- Always remains in Bulu-Bulu's control.
- Never disappears, drops, duplicates, floats, or changes form.
ACTION
0.0–2.5 SECONDS — BREAKTHROUGH
Fight begins mid-battle. Dozens of enemies charge. Bulu-Bulu explodes forward at extreme speed. Three attackers converge simultaneously. He slips through the gap between them, deflecting one attack with the spear shaft, striking another with the blade, and smashing a third with the spear butt. Camera aggressively chases him.
2.5–5.0 SECONDS — REDIRECTION
A new attacker intercepts him. Bulu-Bulu grabs the attacker with his free hand while maintaining spear control and hurls him into advancing enemies. Without stopping, he cuts diagonally through a cluster of attackers, using tight spear movements to redirect weapons and create openings. Camera whip-pans after him.
5.0–7.5 SECONDS — HIGH JUMP
Using collision momentum, Bulu-Bulu launches into a massive high jump above enemies and burning debris. Camera tilts sharply upward. Mid-air, he rotates once while maintaining full spear control, evading attacks from below. He kicks off a burning beam and changes direction laterally.
7.5–10.0 SECONDS — IMPACT RUN
Bulu-Bulu disappears briefly through smoke, then bursts back into view behind another attacker. The spear drives forward, launching the enemy across burning ground. Without slowing, he accelerates toward another incoming wave as the camera races beside him.
END MID-MOTION.
VISUAL RULES
- Dozens of enemies remain active.
- No enemy stands and watches.
- No pauses, poses, or resets.
- Every action directly causes the next.
- Continuous momentum throughout.
- Grounded superhuman speed.
- Realistic destruction and impacts.
CAMERA
Reactive Hollywood action camera. Aggressive pursuit, whip-pans, rapid reframing, lateral acceleration, low/high-angle transitions, natural motion blur.
ENVIRONMENT
Burning battlefield. Dense smoke, fire, embers, damaged earth, collapsing structures, realistic war-zone atmosphere.
SOUND
Footsteps, spear impacts, weapons clashing, fire, debris, bodies hitting ground, battlefield ambience.
NO MUSIC.
One CONTINUOUS FPV shot, no cuts, one unbroken take, EXTREME kinetic energy. The footage NEVER reverses or plays backward. NO slow motion except one half-second beat at the reveal, then an instant snap back to full speed. Photoreal point-blank framing. GHOST CAMERA: only the CAMERA moves — INVISIBLE, no drone, shadow or reflection ever appears in frame, it leaves no wash and casts nothing. The camera ORBITS the two fighters and DIVES in toward each strike, then pulls back into the orbit — always moving, never a static locked-off shot. NO fade to black — the take runs until time runs out, inside the fight.
CINEMATIC CAPTURE: aged retro film footage — vintage anamorphic lens, heavy analog grain, dust, scratches and gate weave, halation glow off the highlights, low faded contrast with milky open shadows, NEVER crushed blacks. Soft warm daylight in a walled courtyard, dappled sun through trees, weathered stone walls. Grade: faded teal and tan, muted desaturated color, warm sun, the worn texture of an old archival film scan — NOT a game engine, NOT animation, NOT modern digital.
@Image1 is the HERO — match his face, build and black sash trousers exactly in every frame. @Image2 is the OPPONENT — match his face and build exactly in every frame. The shot opens ALREADY mid-fight, the camera whipping in a tight ORBIT around the two fighters in a face-off. The camera orbits once, catching the hero's (@Image1) coiled stillness, fingers beckoning, then he EXPLODES forward with a lightning straight punch, the camera DIVING in hard — the lens rushing to his knuckles as they snap past the opponent's (@Image2) jaw, sweat spraying. It whips behind the opponent as he swings a heavy hook; the hero SLIPS it, the fist tearing past the lens, the camera riding the miss out into a fast orbit, ARCING low around their feet as they pivot, then SWEEPING across to the hero's face — jaw set, eyes blazing, a sharp exhaled shout — before DIVING again onto a spinning back-kick into the opponent's ribs, the impact rippling his torso. The opponent staggers; the camera whips around him as he answers with a wild flurry, fists strobing past the lens, the hero weaving between them, each miss whipping the frame. THE REVEAL: half-second slow motion — the hero airborne in a flying kick, leg extended, foot inches from the opponent's chest, both faces taut, sweat suspended in the warm sun — then an INSTANT SNAP back to full speed as the kick CONNECTS, the opponent blasted backward off his feet, the camera driving through the sweat and orbiting tight and low as he crashes to the stone. The hero lands in a low fluid stance over him, chest heaving, and snaps into his guard with a sharp beckoning gesture and a piercing shout, the camera still whipping around him in the haze as more figures rush in from the shaded edges of the courtyard, the runtime running out mid-chaos.
Sound: a driving percussive score with sharp rhythmic hits, the WHIP of limbs cutting air, cloth snapping, bare feet on stone, the hero's (@Image1) sharp high fighting shouts and exhales on each strike, meaty impact thuds — dropping to one held breath of near-silence at the airborne kick, then a MASSIVE concussive impact, then sustained escalating chaos: pounding feet, shouts, more attackers closing, the score surging to the final frame.
Global style lock: aged retro film look — heavy grain, halation, faded low contrast, gate weave, NEVER clean or modern-digital; the camera is never visible, no drone in frame; the camera ORBITS and DIVES, always moving, never a static locked-off shot; @Image1 and @Image2 match their reference sheets exactly in every frame; strikes land OFF-axis past the lens, never dead-center into it; the fight escalates rather than resolves, no directed ending; no distortion, no subtitles, no watermarks.
Scene & Style: A cinematic, high-end modern luxury apartment at night, large floor-to-ceiling windows showing a blurred Hong Kong city skyline with glowing bokeh lights. Moody lighting, cool tones, movie shot style, high detail, 4k.
Characters: [char_s2aj] & [Mia]
Action & Camera: Fast-paced martial arts combat/playful fight scene. [Mia] throws punches and kicks at the man. [char_s2aj] nimbly dodges her attacks and swiftly sneaks a cheek kiss and a quick surprise kiss on her lips during the fight. [Mia] looks shocked and annoyed before continuing her strike. Looping this action pattern 3 times within 20secs. Shooting in one continuous shot, dynamic camera movement, smooth switching between mid-shots and close-up reactions, high action motion. No slowmotion.
“Cinematic anime short film clip, 15 seconds. Busy colorful Asian street market, vibrant vendor stalls everywhere, hanging lanterns, bamboo poles, fruit and vegetable displays, wooden crates, ladders, market umbrellas, food trays, watermelon stacks. Warm afternoon sunlight, very chaotic and colorful environment.
CHARACTER 1 — JACKIE: Light brown skin, sharp expressive Asian features, short messy dark hair, bright quick eyes, lean agile build, early 30s, alert expression. White tank top, dark loose trousers, black cloth shoes. No weapons — uses environment only.
CHARACTER 2 — OPPONENT: Light skin, broad intimidating features, shaved head, mean cold expression, large muscular build. Black jacket, dark trousers, black boots.
AUDIO: Fast paced comedic action music from frame one, building with every exchange, comedic timing on every prop impact, always energetic and fun throughout.
[0:00-0:02] Already mid fight from frame one. Opponent throws a punch — Jackie ducks under it — grabs a nearby bamboo pole from a vendor stall — vaults completely over the opponent using it. Wide shot showing colorful market. Low angle of the vault. SFX: punch whoosh, bamboo pole, vault landing.
[0:02-0:04] Opponent charges — Jackie grabs a wooden crate — throws it — opponent smashes through it — Jackie already moving, grabs a food tray — uses it as a shield deflecting the next punch — spins and hits opponent with the tray. Fast cuts every angle. SFX: crate smash, tray deflect, comedic impact.
[0:04-0:06] Opponent grabs a stool and swings — Jackie ducks — grabs hanging string of lanterns — swings across the market on them — lands on top of a vegetable stall. Opponent charges from below. Wide shot of the swing. Side profile of the landing. SFX: lanterns creaking, stall landing.
[0:06-0:08] Jackie jumps off the stall — uses opponent's momentum against him — redirects him into a stack of fruit — fruit exploding everywhere. Jackie lands using a market umbrella to slow his fall. Very comedic but very skillful. Wide shot of fruit explosion. Close up of Jackie's surprised face. SFX: fruit explosion, comedic music hit.
[0:08-0:10] Opponent back up furious — Jackie grabs a ladder — uses it between both hands spinning it, blocking every strike, opponent can't get past it. Very signature move. Close up of ladder blocking. Side profile of the spinning. SFX: ladder impacts, frustrated opponent.
[0:10-0:12] Opponent grabs one end of the ladder — tug of war — Jackie lets go suddenly — opponent stumbles — Jackie slides under a market table, comes up the other side, kicks the table into the opponent sending him backward. Low angle of the slide. Wide shot of table impact. SFX: table sliding, impact.
[0:12-0:15] HERO MOMENT. Opponent makes one final massive charge — Jackie sidesteps perfectly at the last second — opponent crashes into a massive stack of watermelons — watermelons exploding everywhere in slow motion — Jackie standing calmly in the middle of the explosion, watermelon pieces flying around him — looks at camera — slight embarrassed smile — shrugs — cut to black.
Cinematic cel-shaded 3D anime, semi-realistic CGI, hand-painted textures, Unreal Engine quality, warm afternoon market lighting, vibrant colorful market colors, digital painting feel, hard-edge brushwork, heavy shadows, not cartoon not Disney not Pixar, film grain, premium anime quality, fast paced comedic action music dominant throughout.”
[STYLE + CAMERA + ATMOSPHERE]
Brutal 2026 Viking action cinema with hyper-realistic grounded fight choreography and heavy physics. Single continuous dynamic tracking shot following a massive fat berserker as he rampages through a Viking village, systematically taking down villagers with brutal force and his axe until he reaches his longhouse, discovers his wife's betrayal and explodes in full berserk rage. Anamorphic lens, subtle film grain, cold northern daylight, intense motion blur, stabilized professional handheld camera.
[IMAGE REFERENCES]
Generate consistent character: late-30s massive fat but powerfully built Viking berserker, long wild hair and thick beard, intense wild eyes, covered in Norse tattoos and scars, heavy belly but thick muscular arms. Wears dirty fur cloak over bare scarred chest, leather pants, heavy boots and carries a large blood-stained axe. Exact same appearance, body type, clothing and weapon throughout.
[TIMELINE SECOND BY SECOND]
0-4s: [Continuous Dynamic Tracking Shot] Camera tracks backward through the muddy village center as the berserker advances with growing rage. Several villagers try to stop him; he swings his heavy axe in brutal arcs, cleaving one and throwing another into a longhouse wall with raw power. Heavy impacts, bodies reacting with realistic weight. Camera tight on his massive frame.
4-8s: [Continuous Dynamic Tracking Shot with subtle lateral orbit] Pushes deeper between longhouses. Uses his massive size to bull-rush and crush one against a post, then delivers devastating axe strikes. Fabric, fur and blood move realistically. Growing intensity. Camera orbits slightly while tracking his advance.
8-12s: [Continuous Dynamic Tracking Shot - Building Rage] Approaches his longhouse area, smashing through a door with his body and axe, throws a villager into a fire pit. His movements become more unhinged. Camera tracks forward as destruction builds.
12-15s: [Continuous Dynamic Tracking Shot - Berserk Explosion] Reaches his longhouse and sees his wife with another man. Full berserk rage erupts. He charges with animalistic fury, axe swinging in devastating blows. Camera stays locked on the explosion of violence as he advances, then slowly pulls back revealing the trail of carnage through the village. Seamless continuous take, peak visceral energy.
[STYLE & QUALITY BOOSTERS]
Photorealistic 8K, ultra-detailed skin, fabric, fur, wood and blood textures with realistic sweat, dirt and impact, perfect coherent physics on all strikes, throws, body impacts and environmental destruction, flawless motion blur, stable character consistency, no artifacts, professional blockbuster film look, high dynamic range, intense visceral action, cold northern color palette.
Ultra-realistic cinematic AAA Xianxia action film inspired by high-end martial arts fantasy and Black Myth Wukong combat energy. Ancient stormy mountain temple ruins above the clouds at night, rain, fog, burning incense embers drifting through the air. Erlang Shen and Sun Wukong face each other in silence before battle.
Erlang Shen [Image1] : divine Chinese god warrior, silver-black celestial armor, red waist cloth, third eye glowing faintly, wielding an enormous three-pointed double-edged spear.
Sun Wukong [Image2] : legendary monkey king, dark golden fur details, battle-worn ornate armor, flaming phoenix-red cloth strips, holding the Ruyi Jingu Bang staff.
Both characters use attached face references exactly.
No cartoon style. Pure cinematic realism.
The camera begins with extreme close-ups of wet armor, eyes, fingers gripping weapons, boots sliding across stone. Sudden explosive movement — Wukong [Image2] vanishes using cloud-step speed movement, appearing behind Erlang Shen [Image1] instantly. Erlang blocks without looking using the spear shaft. Sparks explode. Shockwaves ripple through rain puddles.
Rapid wuxia movement, superhuman agility, gravity-defying martial arts, realistic cloth physics, detailed rain simulation, cinematic debris.
Chinese dialogue only:
Sun Wukong [Image2] : “杨戬，你还是老样子。”
Erlang Shen [Image1] : “这一次，我不会让你逃。”
English subtitles burned-in.
CAMERA:
ARRI Alexa 65
Lens: Atlas Orion Anamorphic 40mm and 65mm
Fast handheld + drone-like orbital tracking + speed ramping
LIGHTING:
Cold moonlight mixed with warm fire embers
COLOR GRADING:
Dark teal, silver, gold highlights, high contrast cinematic grading, subtle film grain
NO MUSIC.
Only rain, wind, footsteps, weapon impacts, cloth movement, thunder, breathing.
Ultra-realistic cinematic disaster epic, single continuous shot, no cuts, no montage, no multiple-camera language, documentary-style realism, IMAX-scale clarity, natural film grain, high dynamic range, physically accurate daylight, realistic atmospheric perspective, authentic reflections, volumetric dust and smoke, true-to-life city lighting and environmental behavior. A real modern Chinese megacity in daytime: high-rise towers, elevated roads, broad avenues, plaza spaces, parked and moving cars, street signs, glass facades, all with the look of genuine live-action footage. The shot begins with chaos already underway: a colossal ancient giant python, far larger and more terrifying than a realistic anaconda, is violently rampaging through the city. Its thick muscular body whips between streets and buildings with horrifying speed and weight, crushing cars, smashing roadside structures, wrapping around building facades, lunging at fleeing crowds. People run in panic, vehicles collide, glass shards, concrete dust, torn billboards and debris fill the air. Everything feels like a real catastrophic emergency captured in one uninterrupted shot. Then the sky suddenly shifts. Clouds are torn apart at high speed and a gigantic Chinese dragon dives down from the heavens. This is not a cartoon dragon and has no wings. It is a believable colossal Eastern dragon, a mysterious real biological creature of the natural world, long serpentine body, no wings, powerful limbs, realistic muscles, bones, scale overlap, weathered reflective scales, layered shadows, natural skin tension, believable mass and inertia. As it moves, its body undulates through the air with convincing aerodynamic resistance, whiskers and mane trailing in turbulent airflow, subtle cloud vapor and disturbed mist forming around it. The dragon descends like a guardian over the city, precisely targeting the giant python below. In one continuous shot, the two beasts engage in a brutal close-range fight among skyscrapers and over shattered streets. The python coils, strikes, and thrashes between buildings; the Chinese dragon tears away constriction with its claws, coils around the serpent to overpower it, and finally bites hard into the python's upper body and neck. With overwhelming force, the dragon drags the struggling serpent up from the city ruins and lifts it into the sky. In the final moments, the wingless Chinese dragon powers upward by driving its long body through the air, carrying the still-writhing python away into the distant sky. Below, terrified citizens stop running and stare upward in stunned awe. The ruined city remains filled with smoke, debris, and sunlight cutting through haze, creating a final epic image of a divine dragon saving humanity. Keep the entire piece grounded in real-world physics, realistic destruction, believable scale, terrifying weight, restrained camera behavior, immersive live-action disaster authenticity, not stylized, not fantasy-cartoon, not game-like, absolutely photorealistic.
The film has a retro Hong Kong-style martial arts film aesthetic, with a slight film grain and a warm amber tone.
0-4 seconds: A fixed medium shot from the front, inside an ancient temple hall. Two fairies and a swordsman are depicted in sword-wielding stances. The protagonist in the center holds a long sword horizontally across his chest, while the two on either side each hold a short sword. Both their gazes are fixed ahead. Torches flicker inside the hall, casting orange light and shadow across the stone floor. In the background, the figure of an assassin in a straw hat sways slightly in the shadows. Ambient sounds include the crackling of burning torches and the sound of wind in the distance, intermittently plunging, deep notes from a guzheng (Chinese zither) like heartbeats.
4-8 seconds: A rapid zoom-in to the protagonist's face. The protagonist suddenly opens his eyes wide, pupils slightly contracting—the scene flashes to a close-up of the pupils. The shot cuts back to a medium-long shot at a 45-degree angle from above. All three simultaneously spring up from their kneeling positions, robes billowing and dust rising, weapons drawn and flashing metallic arcs. The protagonist slashes forward with his long sword, the swordsman on the left sweeps his short sword horizontally, and the swordsman on the right delivers a flying kick to the side. The ambient sound abruptly shifts to sharp metallic clanging and the whistling of fabric, accompanied by a rapid, torrential drumbeat.
8-13 seconds: A circular tracking shot moves in a 180-degree arc around the protagonist. The three engage in a chaotic battle with a group of assassins in straw hats, swords clashing and sparks flying. The protagonist spins and delivers a horizontal slash, deflecting a straw hat that spins into the air, the hat spinning twice in slow motion. A warrior on the left flips over an enemy's shoulder to launch a sneak attack from behind, while a warrior on the right parries with his sword hilt and then thrusts back. The fight's pace alternates between fast and slow, with each heavy blow momentarily reduced to 60% speed before returning to normal. The ambient sound consists of dense weapon clashes, shouts of battle, and the vibrations of wooden structures, while the background music shifts to intense pipa tremolo.
13-15 seconds: The camera rapidly pulls back to a wide, frontal angle shot, revealing three warriors standing back-to-back, their weapons pointing in three directions, while all the surrounding straw-hatted assassins fall to the ground. Inside the hall, dust slowly settled, and torchlight beams pierced through the dust, forming Tyndall effect pillars. The scene froze, the sound stopped for a second, and then a deep, lingering chime faded away.
A battle-worn knight clad in dented silver armor and a tattered crimson cape rides a black warhorse through a burning medieval village; ashes drift in the air as flames consume wooden buildings around him, and villagers flee in panic; the camera follows him from behind in a low-angle tracking shot, emphasizing the weight of duty in his solemn posture; the lighting is a dramatic mix of firelight and dusk glow, with warm tones, high contrast, and medium focal length; stylized in a gritty historical epic tone, evoking the atmosphere of a war-torn legend.
Create an ultra-realistic cinematic martial arts fight scene outside a modern Minimarket convenience store on a sunny day. A young East Asian woman with long black hair, wearing a white cropped jacket, a red top, loose dark gray jeans, and white sneakers, fights a young East Asian man with messy black hair, wearing a black tactical jacket, a dark green shirt, black cargo pants, and black combat boots. The fight features fast martial arts combinations, powerful high kicks, spinning kicks, acrobatic jumps, dodges, blocks, and realistic counterattacks. The woman demonstrates impressive fighting skills, knocking the man off balance and sending him crashing against a concrete wall. Bright blue sky, fluffy white clouds, green trees, colorful convenience store signage, wet reflective pavement, yellow road markings, realistic shadows, dynamic handheld camera movements, low-angle action shots, wide shots, close-up facial expressions, natural motion blur, dramatic impact moments, physically believable body movements, detailed clothing textures, realistic human anatomy, photorealistic skin, cinematic color grading, 4K resolution, vertical 9:16 aspect ratio, high-end action movie cinematography. No subtitles, no watermarks, no on-screen text.
[0–3s]
Camera: Rapid push-in from a close-up of the blade's glint, while the camera rotates clockwise at 180°/s, cutting into the silhouette of a bamboo hat (douli).
Visuals: Close-up of ink-wash blade light, particle sparks bursting outward; the douli silhouette emerges from billowing ink clouds.
Effects: Motion blur, ink brushstroke trails, particle effects.
[3–6s]
Camera: High-speed 360° orbit centered on the character (270°/s), following the trajectory of the sword swing.
Visuals: Medium shot of the character swinging the blade, robes billowing, ink brushstrokes flowing with the motion.
Effects: Dynamic ink smoke, motion blur on the robes, visual drag caused by the rotation.
[6–9s]
Camera: High-speed lateral tracking shot (left → right), while the camera tilts and rotates counterclockwise at 90°/s, locked onto the blade's path.
Visuals: The blade's trajectory line, particle sparks bursting outward, ink brushstroke trails.
Effects: Motion blur, particle effects, ink ghosting trails.
[9–12s]
Camera: Low-angle upward shot with a fast crane-up, while the camera spirals at 225°/s, rising from the ink marks on the ground to the character's upper body.
Visuals: Ink marks on the ground, the character mid-swing (upper body), rolling ink clouds in the background.
Effects: Dynamic ink clouds, motion blur during the rise, ink-colored vortex created by the rotation.
[12–15s]
Camera: Slow-motion close-up, with the camera rotating slowly clockwise at 15°/s as the blade light freezes in place.
Visuals: Close-up of the blade light, crisp and clear particle effects, the character's gaze beneath the shadow of the douli (optional).
Effects: Slow motion, particles dissipating, subtle camera shake.
A cinematic action fight scene inside an old, crowded Pakistani passenger train compartment. A young woman with long brown hair, wearing a blue denim jacket and loose white pants, fights multiple men dressed in black. She performs fast, realistic martial-arts moves, dodging punches and delivering powerful kicks and strikes. The fight moves through the narrow aisle between old worn-out blue train seats. Dynamic handheld camera movement, low-angle shots, wide shots and close-ups, realistic body movement, natural motion blur, dramatic physical choreography, intense expressions, dusty vintage train interior, daylight coming through the windows, gritty realistic atmosphere, cinematic color grading, shallow depth of field, highly detailed, photorealistic, realistic skin texture, Hollywood-style action cinematography, 35mm lens, high contrast, 4K, smooth motion, no text, no subtitles, no watermark.”
Negative Prompt
“cartoon, anime, CGI look, unrealistic anatomy, extra fingers, extra limbs, distorted face, duplicate people, floating objects, unnatural movement, excessive blood, gore, blurry faces, low quality, oversaturated colors, text, subtitles, watermark, logo.”
SCENE 1
REFERENCES:
Image 1 = Ilya
Image 2 = Ilya’s Spear
Image 3 = Rica
Image 4 = Rica’s Spear
CHARACTER LOCK:
Image 1 remains Ilya’s exact identity throughout.
Image 3 remains Rica’s exact identity throughout.
Never swap, merge, replace, or mix their identities. Keep both fighters correctly separated.
STYLE:
GOD-LEVEL PHOTOREALISTIC LIVE-ACTION COMBAT
EXTREME SPEED • CONTINUOUS MOTION • AGGRESSIVE CAMERA
NO STATIC MOMENTS • NO NORMAL RUNNING
ACTION
0–2 SECONDS:
Ilya and Rica face each other. Purple electricity erupts across Ilya and his spear. Red electricity erupts across Rica and her spear. Both EXPLODE forward simultaneously with extreme speed, instantly erasing the distance. Camera flies diagonally through the battlefield as they tear past from opposite directions, violently shaking the frame from air displacement. Camera continues forward instead of following.
2–4 SECONDS:
Both instantly reverse and collide again from different directions. Ilya’s spear creates multiple changing purple trajectories while Rica’s spear simultaneously creates red trajectories. Both attack while already moving and changing direction. Camera intercepts from above, dives diagonally through the electrical wake, then violently redirects toward the next clash.
4–6 SECONDS:
Continuous high-speed exchanges. Ilya flashes across one side while Rica attacks from another. Their bodies cross → separate → instantly reverse → cross again. Purple and red electrical trails tear through the battlefield. Camera races across Rica’s spear path, banks away from Ilya’s spear, then shoots upward through the electrical collision.
6–8 SECONDS:
The fight accelerates further. Both attack while moving away and changing direction. Spears never stop. Purple and red electricity repeatedly intersect across different locations. Camera arrives low beneath their crossing point as both blast overhead, then instantly shoots sideways, rises, and dives toward another clash.
8–10 SECONDS:
FASTEST CONTINUOUS BURST. Ilya and Rica repeatedly tear across the battlefield, changing direction instantly. Neither spear stops. Purple and red attack lines continuously overlap, collide, separate, and reappear from new directions. Camera constantly changes route: side interception → violent direction change → rise above red trajectory → break away from Ilya’s purple spear → diagonal dive into the next clash.
CAMERA:
Never static. Never locked to one direction. Never simply follows behind either fighter. Every extreme pass, electrical collision, acceleration, and impact causes realistic directional jolts and rapid vibration.
SOUND:
NO DIALOGUE. NO MUSIC.
Only extreme wind, air displacement, rapid spear movement, aggressive purple electricity, violent red electricity, weapon impacts, electrical collisions, and environmental reactions.
ABSOLUTE RULES:
NO IDENTITY SWAP.
NO CHARACTER MERGING.
NO NORMAL SPRINTING.
NO TELEPORTATION.
NO DEAD TIME.
NO STATIC CAMERA.
NO SLOW ATTACKS.
SPEARS ALWAYS IN MOTION.
BOTH FIGHTERS MOVE SIMULTANEOUSLY.
PHOTOREALISTIC LIVE-ACTION ONLY.
Cinematic dark fantasy wuxia action scene, low angle dynamic tracking shot. A male warrior in black martial robes and arm guards stands in a tall grass field under an overcast gray sky. He spins dynamically, slashing a glowing fiery sword that leaves arc trails of bright orange fire and sparks. A massive swarm of black shadow beasts charges across the field. He triggers a powerful ground shockwave that blasts dirt, smoke, and embers into the air. A circular matrix of glowing red target reticles and energy lasers shoots outward into the monster swarm. The warrior then rests his blazing sword on his shoulder and walks forward in slow motion through floating embers, smoke, and dissolving black ash. Photorealistic VFX, motion blur, 8k resolution, hyper-detailed.
A single continuous 20-second cinematic long take set inside a lavish luxury hotel ballroom at night. Massive floor-to-ceiling windows reveal a neon city skyline, elegant chandeliers hang overhead, polished black marble floors reflect the lights, overturned tables and scattered glass create a dangerous environment. The scene feels glamorous and sophisticated, but suddenly erupts into an extremely intense, fast-paced hand-to-hand fight.
[0:00–0:03] Medium tracking shot. A sophisticated woman with sharp green eyes, bold dark brows, and deep red lips walks calmly through the ballroom. She wears an adorable feminine outfit: a fitted pastel-pink cardigan, white pleated mini skirt, delicate ribbon at the collar, white ankle socks, and elegant Mary Jane shoes. She looks completely composed and graceful. Without warning, a large muscular assassin in a dark tactical outfit charges toward her from behind.
[0:03–0:06] She instantly turns and blocks his first punch. He throws a rapid combination of heavy strikes. She narrowly dodges each one, ducking, twisting and stepping around him with precise martial-arts movement. Camera circles tightly around them, creating intense kinetic energy. He grabs her arm and attempts to throw her — she reverses the movement and slams him hard onto a marble table.
[0:06–0:09] Before he can recover, he violently sweeps her legs. She hits the floor, rolls underneath a punch, springs back to her feet and immediately launches a brutal counterattack. Rapid punches, elbows, knee strikes, blocks and evasive movements. Her cute skirt and cardigan move naturally with every movement. Realistic weight, momentum and body mechanics.
[0:09–0:12] The assassin grabs her and drives her backward through a decorative glass partition. Camera follows through the impact in one fluid movement. Glass explodes around them. She lands, rolls across the marble floor and immediately rises. Her expression becomes cold and determined. The assassin charges again.
[0:12–0:16] EXTREMELY INTENSE FIGHT SEQUENCE. Continuous rapid choreography with no pause: she parries a punch, strikes his ribs, ducks under a hook, counters with a spinning kick, blocks a heavy strike, grabs his arm and uses his momentum to throw him across the room. He crashes into a wooden table. He immediately gets back up and rushes her. She uses the environment, stepping onto a fallen chair and launching herself into a powerful flying strike. Camera rapidly tracks, whips around and crashes close during every major impact. Realistic motion blur, powerful sound-implied impacts, debris flying through the air.
[0:16–0:18] The assassin attempts one final desperate attack. She catches his arm, pivots with perfect timing and executes a devastating Judo-style throw, sending him crashing onto the marble floor. The impact sends loose glass fragments sliding across the reflective surface. She remains standing, breathing heavily but composed.
[0:18–0:20] Extreme close-up. Her sharp green eyes lock onto the defeated attacker. A tiny strand of hair falls across her face. She calmly fixes her pink cardigan, straightens her pleated skirt and adjusts her ribbon as if nothing happened. She turns and walks away gracefully through the ruined ballroom while city lights shimmer behind her. Camera slowly pushes toward the shattered glass and fades to black.
STYLE: Photorealistic, ultra-detailed 4K cinematic action film, extremely intense hand-to-hand combat, sophisticated feminine aesthetic contrasted with brutal choreography, realistic martial-arts technique, physically accurate body mechanics, realistic impacts and momentum, natural cloth and hair movement, cinematic motion blur, dynamic handheld camera, continuous uninterrupted camera movement, rapid tracking shots, whip pans, dramatic close-ups, shallow depth of field, realistic glass destruction, practical lighting, warm chandelier light mixed with cool blue city light, glossy reflections, high contrast, premium Hollywood action cinematography, relentless pacing, no cuts, one continuous shot.
For the target video, at 0.00 seconds into the target video, <Picture 1> (from [Shot 1]) is fully referenced.
integrated_multimodal_description: [Shot 1] Live-action, cinematic realism, epic fantasy war climax, hyper-real and tragic. The same red-haired female warrior from <Picture 1> is fully preserved in face, hair, body proportions, dual swords, armor, burning battlefield, skulls, spears, smoke, embers, sparks, and infernal war atmosphere. IMAX 70mm film grain, inspired by Nolan cinematic aesthetics, photorealistic natural color grading with charcoal black, umber, dark ochre, ash gray, and pale amber interwoven. The whole video is one continuous shot with no cuts. The camera stays low and close, retreating under pressure while also making slight side drift, subtle breathing-like push-pull, one short impact disturbance, a curved move around her front-side during the breakdown, and a low dipping follow into the fall. From 00:00.000 to 00:04.500, she charges straight toward the lens at high speed, chest angled slightly toward camera-left, head forward, eyes locked ahead, mouth slightly open, breathing hard, shoulders and core tight. One sword drives low at frame-right, the other opens wider at frame-left. Hair, straps, armor, cloth, and blades all carry strong secondary motion. Embers cross the lens and flames boil behind her. From 00:04.500 to 00:06.200, a heavy arrow blasts in from the off-screen right-front side and hits high in her right upper torso near the shoulder and chest. The impact is like a hammer: right shoulder jerks back, chest contracts, head and neck snap slightly rearward, and her breath is cut off. The camera gives one short force reaction. From 00:06.200 to 00:10.500, she does not fall immediately. She forces half a step to one and a half more steps forward while losing control: upper body tipping backward, lower body still trying to drive ahead, compensation steps failing, pain and disbelief visible in her face. The camera arcs around her front-side to watch the breakdown. One sword slips free first and is thrown or dropped; the second weapon also loses control and falls out of its attack line. Hair, shoulders, arms, armor, leather, and weapons all lag with visible inertia while sparks and fire whip around her. From 00:10.500 to 00:13.200, the collapse fully breaks through in layers: upper body whips farther back, chest and shoulder are dragged open by the wound, legs try one last catch and fail, knees buckle, and she slams backward or diagonally backward with heavy force. The camera follows the fall with a low dipping angled movement. From 00:13.200 to 00:15.000, her body and weapons hit the ground with one deep heavy impact, throwing up dust, embers, sparks, and debris. The camera stays low with a small drifting or sinking continuation, holding her as a fallen legend rather than a weak victim. Fire and battlefield ash keep moving to the end.
overall_soundscape: Heavy running footfalls, armor and weapon movement, wind, ember crackle, distant fire roar, arrow whistle, brutal impact into armor and body, broken breath, stumbling loss of control, weapon drops, and one deep final body-and-metal crash.
non_diegetic_music: N/A
One continuous shot. The young Korean man, same matte-black combat suit, same blue-glowing gravity blade, same three black armored robots, and the same rain-soaked neon alley remain unchanged throughout. No teleporting, no position resets, no sudden costume/weapon changes, no jump cuts.
0–3 sec: The man walks slowly forward through the neon alley while the three robots approach him from ahead. Camera: smooth backward tracking at chest height, keeping him centered. Rain falls naturally and blue blade light reflects on the wet ground.
3–6 sec: Without cutting, the robots charge. He stops, plants his left foot, and swings the blade horizontally. A blue gravity wave expands directly from the blade toward the closest robot. Camera: smoothly circles from his front-left to his side, following the same movement.
6–9 sec: The gravity wave hits the first robot and throws it sideways into the wall; the other two lose balance and become partially weightless. The man remains exactly where he finished the swing. Camera: continues the same orbit and reveals the robots floating around him.
9–12 sec: One robot recovers and jumps toward him. He raises the same blade upward and reverses its gravity, pulling the attacker sharply toward the ceiling while the other robot remains suspended behind him. Camera: smoothly tilts upward with the robot, then follows the blade back down to the hero.
12–15 sec: The man lowers the blade and releases the gravity field. The floating robots and loose debris fall naturally to the ground around him. He remains standing in the same position as rain continues falling. Camera: slowly pushes toward his face as the blue glow of the blade fades slightly.
Visual style: continuous cinematic sci-fi action, anime-inspired but original, realistic physics, detailed mechanical robots, rain, neon reflections, volumetric fog, natural motion blur and smooth camera choreography. No cuts, no scene changes, no teleportation, no sudden jumps in character position.
Medieval war epic, montage. A young knight crosses fire on the night the royal city falls, finds the heir and escorts them out of the castle. Besieged stone castle, burning throne hall, walls under arrow fire, dark tunnels, forest edge before dawn. Armour, banners, stone bridges, warhorses. Iron grey, flame orange-red, dark gold, deep forest green. Thick smoke, torches and fire throwing jumping highlights, cold light on stone and steel, ash and dust in the air. Third-person on the hero, cross-cut with the wider battle. Big wides, fire close-ups, fast intercutting, slow-motion charge, tight handheld in cramped spaces. War drums, horns, metal impacts, fire, horses screaming, low choral voices.
30-second ultra-realistic cinematic fight scene, 16:9. Two adult male fighters face off inside a rain-soaked abandoned warehouse at night. Keep the SAME two men, clothing, and appearances throughout. Intense choreographed action with realistic physics. No injuries, blood, wounds, or graphic violence.
[0–5s] Wide shot. Fighters circle cautiously beneath flickering lights. Heavy rain pounds the roof, distant thunder rumbles, shoes scrape the concrete. Both breathe heavily.
[5–10s] Fighter one throws a controlled punch. Fighter two ducks and counters with a blocked strike. Realistic whooshes, impacts, sharp grunts, footsteps, and clothing movement.
[10–15s] Rapid punches and defensive blocks. Each fighter makes natural effort sounds—short grunts, forceful exhales, heavy breathing. Punches create crisp impact sounds without exaggerated bone-cracking.
[15–20s] Fighter two attempts a sweeping kick. Fighter one jumps backward. Loud shoe scrape, kick whoosh, landing thud, rain and thunder continue.
[20–25s] Fast dodges, blocks, spins, and controlled martial-arts combinations. Dynamic camera circles them. Realistic impacts, grunts, breathing, footsteps, and environmental echoes.
[25–30s] Both attempt a final strike but stop inches apart. Silence briefly cuts through the action except for their heavy breathing and rain. They lower their fists and step backward as thunder rolls.
Photorealistic humans, realistic anatomy, natural physics, cinematic lighting and synchronized audio. No music. No dialogue. No injuries, blood, bruises, weapons, gore, death, broken limbs, distorted anatomy, duplicated characters, or morphing.
A structured 15-second high-intensity action duel with explicit hard cuts. [SHOT 1: 0s-4s] Dissolving smoke on the lantern-lit street reveals the dark warlord unharmed, standing tall as his eyes glow with fierce red energy, raising his black dragon staff. [HARD CUT TO SHOT 2: 4s-8s] Dynamic low-angle tracking shot of the beautiful white-robed female swordswoman sprinting forward at lightning speed across the stone cobblestones, her glowing silver sword leaving a trail of blue light, white silk dress fluttering. [HARD CUT TO SHOT 3: 8s-15s] Intense close-quarters martial arts collision! The female swordswoman leaps and strikes her silver sword violently against the warlord's dragon staff, unleashing a blinding explosion of bright sparks and shockwaves. They rapidly exchange three furious parries—a spin kick, a staff thrust, and a sword slash—ending in a fierce weapon lock-up with intense face-to-face eye contact under floating red lanterns.
[PROSE & IN-LINE SPATIAL & GLOBAL LOCK]: [0s-1s] In a supermarket aisle, Kenji in an open suit chops a Katana; at 0.25s Chinese housewife Mei-Ling cradling toddler Wei in her left arm blocks with a cabbage in her right hand; at 0.5s the blade chops the cabbage; at 0.75s she thrusts half-cabbage into his face. [1s-2s] At 1.25s he slashes; at 1.5s Mei-Ling high-blocks; at 1.75s she swings Wei's body so Wei's sneaker kicks Kenji's chin. [2s-3s] At 2.5s Mei-Ling parries his blade with leeks, exploding green mist. [3s-4s] At 3.25s she rams a shopping cart into Kenji's shins; at 3.75s Kenji winces, speaking native Japanese: "Kuso!" [4s-5s] At 4.75s Mei-Ling side-steps and drives a Wing Chun punch into his ribs. [5s-6s] At 5.75s Mei-Ling ducks while Wei's small foot taps Kenji's nose. [6s-7s] At 6.75s Mei-Ling slaps a carrot tray across Kenji's cheek. [7s-8s] At 7.5s she front-kicks Kenji's chest. [8s-9s] At 8.75s Kenji's blade lodges between cart steel bars. [9s-10s] At 9.5s Mei-Ling twists the cart, snapping the Katana blade with a metallic ring. [10s-11s] At 10.25s Mei-Ling delivers an open-palm chop to his neck; at 10.75s she shouts, speaking fluent native Mandarin Chinese: "Beyao darao wo mai cai!" [11s-12s] At 11.5s Wei's sneaker kicks Kenji's eye. [12s-13s] At 12.25s Mei-Ling smashes a watermelon onto Kenji's head, bursting red juice. [13s-14s] At 13.25s she backfists his temple, causing Kenji to drop his broken hilt. [14s-15s] At 14.0s Kenji collapses onto the terrazzo floor; at 15.0s Mei-Ling adjusts Wei and picks up her groceries. [FG: Flying cabbage & water mist | Mid: Mei-Ling & Kenji combat | BG: Refrigerated produce shelves]
[ACTING & LIGHTING SCIENCE]: Jackie Chan action comedy drama. Authentic micro-expressions, fierce maternal protective eyes, unforced kinetics. Affirmative skin physics: Bright natural fair ivory skin tone, translucent epidermal pores, unpolished realism, subtle sweat. Bright 5000K LED lighting clashing with 3000K warm accent spotlight on white terrazzo floor. Uncompressed 32-bit float audio mix, clear vocal headroom, heavy sub-bass impact weight. [COLOR GRADE LOCK]: Primary Triadic Separation of Crimson, Emerald, and White.
[CAMERA SCIENCE & KINETIC PHYSICS]: [SUBJECT MOTION: Level 4 hyper-velocity 250ms sub-second martial arts pairings, toddler sneaker kicks, cabbage shield, sword snap] [CAMERA MOTION: Low-angle pursuit tracking pan, 45-degree staccato shutter angle for blur-free high-velocity combat. Smooth fluid 24fps cinema motion, natural shutter, continuous fluid physics]. Clean wall architecture. Strict 100% Euclidean architecture: double-row stainless steel shelves, clean white terrazzo floor geometry. Fast f/1.4 Anamorphic prime lens, creamy background bokeh blur. Edge-to-edge optical clarity, 16:9 aspect ratio. Real-time 1.0x. Strict single-character weapon ownership lock: Mei-Ling strictly uses martial arts & props, Kenji strictly holds Katana, zero weapon swapping, zero blade morphing.
A dynamic 9-second cinematic martial arts fight scene inside a large indoor training gymnasium with a smooth teal-green mat floor, high industrial ceiling with metal beams, an American flag hanging in the background, and rows of spectators in white t-shirts watching from the sidelines. Soft dramatic lighting with slight haze.
A muscular Asian man with short dark hair, wearing an olive-green military-style button-up shirt, matching cargo pants, and black combat boots, aggressively attacks with intense facial expressions and open-mouthed shouts. A graceful young woman with long blonde wavy hair, dressed in a light blue traditional Chinese martial arts tunic with black frog buttons, white loose pants, and white sneakers, remains calm and precise, using fluid Tai Chi/Kung Fu techniques.
Sequence:
The man leaps high into the air toward the woman with clenched fists and a fierce shout.
He delivers a powerful mid-air high kick; she blocks and redirects it.
He drops low to the mat in a predatory stance, crawling forward with intense focus.
She settles into a strong defensive horse stance, both palms extended forward in a guarding posture.
Close-up of their hands clashing as he pushes aggressively and she redirects his force.
She grabs his raised boot and flips him backward; he crashes onto the mat on his back with legs in the air.
He quickly recovers into a ready fighting stance, fists raised, face contorted in determination.
He jumps again with arms raised for another strike as her white-clad leg counters.
Final shot: the woman stands alone in a balanced, elegant martial arts pose on the green mat, hands raised in a calm, powerful guard, looking directly forward with quiet confidence while the man is defeated off-screen.
Smooth tracking and dynamic camera angles, realistic physics, high-detail clothing movement, cinematic action style, sharp focus, 24fps feel.
Cinematic 15-second storyboard, vertical/portrait aspect ratio, ultra-detailed anime-realism hybrid style, epic wuxia fantasy action.
Character (Heroine – exact match to reference): Beautiful young woman with waist-length flowing pure white hair, wearing an ornate white horned mask with sharp upward horns covering the upper face, revealing only lips and chin. Form-fitting white battle dress with open midriff, layered sheer white fabric, white arm wraps, thigh high white bandages, and white heeled sandals. Dual-wielding: right hand a long white sword with wooden handle held high, left hand a pure white blade held low. Pale skin, athletic build, dynamic combat stance.
Setting: Wet circular stone platform high on misty mountain cliffs, towering jagged green-gray rock peaks in background, soft overcast light, light fog, water droplets on the ground.
Opponent: Dark mirror counterpart – tall male warrior in black tattered armor with crimson accents, dual black blades, red glowing eyes under a horned black mask.
15-second timed sequence (storyboard panels):
0–3s
Wide establishing shot → medium shot.
Heroine stands in ready pose on the wet platform, white hair and fabric gently moving in the wind. Opponent steps into frame from the mist. Both raise their dual blades. Tension builds, slight camera push-in.
3–6s
Fast cut to close-up of eyes through masks → full body dynamic exchange.
Heroine launches forward with a spinning slash using the wooden-handled sword. Opponent blocks and counters with a low sweep. Sparks and white energy trails fly. Quick side-tracking camera follows the clash.
6–10s
Mid-air aerial combo.
Heroine leaps high, white hair and dress flowing dramatically, dual swords crossing in an X-slash. Opponent jumps to meet her. Multiple rapid blade clashes in the air, white and black energy trails intertwining. Slow-motion moment as blades lock, then they separate and land hard on the platform, water splashing.
10–13s
Ground power clash.
Both charge, blades colliding in a massive impact. Camera circles 180° around them. Heroine uses the momentum to spin low and strike upward. Opponent staggers back. Intense motion blur on the swords, glowing white energy cracks the stone floor slightly.
13–15s
Final decisive moment.
Heroine performs a graceful yet deadly finishing dual-sword thrust, white energy exploding outward. Opponent is pushed back to the edge of the platform. Freeze-frame on her powerful pose (matching the reference image energy), white hair and fabric dramatically billowing, mountains in background, soft mist swirling. Fade to black with lingering energy particles.
Style notes for every panel:
Photorealistic skin and fabric details, cinematic lighting with soft volumetric fog, high-speed action clarity, elegant feminine power, no blood, pure fantasy martial arts energy. Consistent character design throughout.
Single continuous shot, one take no cuts, cinematic oner, photorealistic,
8K ultra-high-definition, hyperdetailed, 35mm film quality, professional
color grading, sharp focus, fine film grain, depth of field mastery,
sweeping crane-to-tracking fluidity, 180° shutter motion blur.
SUBJECT: an original design — a colossal humanoid-serpentine monster made
entirely of churning black seawater and glowing bioluminescent teal veins
threading through its translucent liquid form, no solid skin, constantly
reforming and sloughing water as it moves, taller than the tallest warship
mast, faceless but for two pale glowing eyes deep within the water-mass.
Same design throughout, no drift. Setting: a night ocean battle, a fleet of
a dozen wooden ironclad-era warships scattered across dark heaving water,
cannons flashing, torches lit along the decks, a storm gathering overhead.
COLOR 60:30:10 — 60% deep black ocean and storm-grey sky / 30% cold teal
bioluminescent glow through the monster's water-body and reflected on the
waves / 10% accent of hot orange cannon-fire and burning ship timber.
PHYSICS: the monster's body behaves as real seawater under immense force,
surging, collapsing, reforming, throwing spray and foam as it moves,
ships rock and are swamped in its wake, cannonballs punch through and
splash out the far side of its liquid mass with no effect, wood splinters
and fire scatters on impact.
0–4s — THE ERUPTION, real-time 24fps: the ocean surface bulges and cracks
open as the monster's colossal head and shoulders explode upward between
two warships, a wall of water and foam crashing outward, capsizing one ship's
rigging. Wide low-angle shot craning up to take in its full height against
the storm sky.
4–5s — THE VOLLEY: sailors scramble, cannons on three ships fire in unison,
muzzle flashes strobing orange in the dark. Time begins to violently ramp as
the cannonballs streak toward the creature.
5–10s — THE IMPACT, slow motion 120fps: near standstill as the cannonballs
punch clean through the monster's translucent water-body, water splashing
outward around each ball in slow arcs, teal bioluminescent veins rippling
and scattering like ink in water at each impact, utterly unharmed, its pale
eyes narrowing with something like fury. Droplets and foam hang suspended,
lit by the orange muzzle flashes and storm lightning.
10–15s — THE RECKONING, snap back to 24fps: time cracks back, the monster
roars a deep resonant water-choked bellow and brings one massive liquid arm
down across the nearest warship, shattering its hull and masts in a wave of
water and splinters, the camera pulling into a vast wide as the fleet scatters
and burning wreckage drifts on the churning black sea.
AUDIO (no music): storm wind, waves crashing against hulls, distant thunder.
At the eruption, a massive roar of displaced water. At the volley, sharp
cannon booms and shouted orders. During slow-mo, sound drops into a deep
muffled underwater pulse, only a slow heartbeat-like resonance. At 10s a
crushing wave-roar and splintering wood as full sound crashes back.
FORBIDDEN: no franchise or recognizable characters, no anime or cel-shaded
style, no design drift, no solid humanoid skin or face, no plastic CGI gloss,
no cartoon proportions, no readable text or logos, no desaturated grading,
no warping geometry outside the creature's intended liquid form.
A dark anime exorcist with long black hair, golden prayer beads floating around his neck, spear wrapped in glowing talismans, focused eyes and calm deadly posture + 0–4s: stands inside an abandoned temple lit by daylight beams as shadow yokai crawl from walls and ceilings, camera slowly pushes through dust toward his spear;
4–10s: he spins into combat, spear cutting arcs of golden fire, talismans exploding into sacred symbols, yokai disintegrating into black smoke, camera orbits through the fight, whip pans between enemies, crash zooms on spear impacts, MAPPA-style black-and-gold impact frames filling the screen;
10–15s: he drives the spear into the floor, releasing a massive circular purification wave that vaporizes every monster, temple bells ring silently as dust clears, camera pulls back to show him alone in the restored sunlight
Ancient Japanese temple, broken statues, floating dust, black smoke, golden holy VFX, supernatural combat climax. epic anime
Masterpiece quality, 6-second video of a highly skilled fighter traversing rough terrain while engaged in a brutal, high-intensity fight. The character performs fast, powerful martial arts combos including spinning kicks, flying knees, aerial flips, wall runs, and savage takedowns. Strong realistic physics with heavy impact and momentum. Dynamic camera work with tracking shots, circling camera, and dramatic low angles. Cinematic lighting, sharp motion clarity, intense and raw action, no cuts, smooth and powerful choreography, best quality, 6-second video.
Shot on ARRI ALEXA 65, 28mm anamorphic. 2.39:1. Muted colors, wuxia desaturation, cold jade-white light bleeding into warm gold. Crushed blacks. Film grain. [Use for the hero character] face, dark hair in double bun with loose strands, white and gold ornate martial arts outfit, gold-embroidered cropped top, wide linen pants, white sneakers, spiked golden gauntlets appearance only. 0.0s–3.0s: Extreme low-angle wide shot the warrior stands alone at the center of Dracula's castle grand hall, massive gothic stone columns rising on both sides, cracked stained glass windows bleeding moonlight, chandeliers swaying above. Skeleton warrior silhouettes emerge from the shadows all around her, hollow eye sockets glowing faint green. She sinks deeper into her fighting stance, fists raised, white smoke curling from her gauntlets, eyes locked forward with cold precision. Camera pushes in slow and low from wide to medium. Audio: distant thunder outside castle walls, chains dragging on stone floor, hollow wind through broken arches. 3.1s–6.0s: TIME REMAP: sudden speed ramp into real-time she explodes forward, golden energy detonating from her fist on impact with the first skeleton warrior, bones shattering and scattering across the stone floor. She pivots into a low spinning kick, sweeping three more skeletons off their feet simultaneously. Camera races beside her in aggressive handheld tracking, barely keeping up. Maintain: dark double bun hair, white gold-embroidered outfit, spiked golden gauntlets, white sneakers throughout all frames. Audio: bone shattering like dry wood, deep energy pulse on each strike, boots crashing on stone, cathedral reverb swallowing every impact. 6.1s–9.5s: Orbit shot camera circles her at medium distance as skeleton warriors attack from every angle simultaneously. She deflects with her forearms, white smoke and golden light bursting from contact points, spins through two attackers at once, uses a fallen column to vault upward and drive both fists downward into a cluster of skeletons below. Green spectral energy meets white fire where she lands. Audio: roaring battle echo in cathedral reverb, cracking bone percussion, distant organ chord vibrating the walls. 9.6s–12.5s: TIME REMAP: ultra slow-motion she delivers the final devastating uppercut, golden chi energy erupting upward from her fist in a column of white light, a shockwave radiating outward across the hall floor, every remaining skeleton warrior blasted backward simultaneously, bones and armor fragments suspended frozen in the air around her. White smoke freezes mid-curl. Loose hair strands suspended in zero gravity. Camera catches debris hitting the lens. Audio: deep subsonic pulse, silence stretching like glass before shattering. 12.6s–15.0s: Slow crane pull-back and rise she stands alone, breathing controlled, fists lowering slowly, white smoke dissipating from her knuckles, surrounded by scattered bones and extinguished torches. Gothic hall stretches impossibly long behind her. One candle still burning in the darkness far above. Camera rises slowly revealing the full scale of devastation. Audio: single candle flame crackling, wind through broken stone, silence. STYLE NOTES: Wuxia meets gothic dark fantasy. Light starts cold jade-white and deep shadow then fractures into gold and white energy bursts during combat, then collapses back to cold silence. Her white outfit becomes a blade of light cutting through the darkness of the castle. The camera worships her precision. Every strike feels inevitable. Single continuous feel, camera chases never leads. Muted colors, wuxia-gothic grading, deep shadow detail preserved. 4K Ultra HD, Rich details, Sharp clarity, Cinematic texture, No blur, No ghosting, No flickering, Stable picture.
A battle-hardened medieval warrior with heavy scratched armor, cape torn, blood across his face, gripping a flaming sword with both hands
Charges into a massive battlefield against monstrous creatures, blocking giant attacks and carving through enemies while catapults explode and fire spreads across the battlefield
Dark fantasy warzone under storm clouds with burning siege towers, mud, ash and thousands of soldiers fighting in the distance
Begins with wide IMAX-style battlefield reveal, slow push-in on the warrior, sudden speed ramp into combat, handheld tracking through brutal clashes, whip pans following sword swings, slow-motion sparks and blood during impacts, lightning illuminating the battlefield, ending with the warrior standing atop a mountain of defeated monsters while rain extinguishes the surrounding flames, camera slowly orbiting his victorious silhouette
A medieval knight in heavy armor, scratched and battle-worn, gripping a longsword with both hands, breathing heavy but steady
Clashes against multiple enemies, blocking blows with brute force, pushing forward with powerful strikes and shield bashes
Muddy battlefield under stormy sky, rain falling, chaos all around
Low-angle tracking shot pushing into combat, weight emphasized in movement, whip pans connecting each clash, slow motion sparks on sword impacts, dramatic lightning flashes
A dramatic aerial wide shot of a vast, dry mountain range with rugged cliffs and dusty terrain, resembling an ancient battlefield. The environment is harsh and barren, with wind blowing sand through the air under a golden sunset sky. From high above, multiple warriors descend using parachutes, landing across the rocky slopes. They are dressed in detailed historical battle armor from 400–500 years ago — leather and metal armor, capes flowing, helmets, and traditional warrior outfits. From the opposite side, another group of enemy soldiers also descends with parachutes, landing aggressively across the mountain terrain. As soon as both sides touch the ground, they quickly detach their parachutes and draw swords. Without hesitation, they charge toward each other across the dusty landscape. The camera shifts to dynamic close-up shots as warriors clash swords, sparks flying on impact, dust rising from the ground, and intense combat movements. The sound of metal clashes echoes through the mountains. Slow-motion moment: two warriors leap toward each other mid-air, swords colliding dramatically. Additional cinematic detail: torn flags fluttering on the mountain, distant echoes, and wind adding intensity to the scene. Final shot: a wide-angle view of the chaotic sword battle spreading across the mountain as the sun sets behind the peaks. Style: ultra-realistic, cinematic, historical action film, dramatic lighting, 4K quality Camera: aerial shots, fast cuts, slow motion during sword clash, dynamic tracking Mood: intense, epic, historical battle.
ultra realistic cinematic scene, large group of ninjas engaged in an intense battle inside a dense bamboo forest at night, heavy rain pouring, thick mist and fog drifting between tall bamboo stalks, dramatic samurai sword fight, multiple katana clash...
Live action cinematic action film. A breathtakingly beautiful female warrior in her late 20s, with long dark braided hair and battle-worn leather and steel armor, faces off against a massive armored villain on a rain-soaked rooftop at night. She moves with lightning speed — a flurry of precise sword strikes creating arcing trails of steel. Her eyes burn with fierce determination. Slow-motion close-up of their blades locking. Rain splatters dramatically off the metal. Shot in the style of a high-budget Hollywood action film — anamorphic lens flare, shallow depth of field, moody blue-orange color grade. Camera does a fast 360-degree orbit during the clash. 4K ultra-realistic.
Epic sword fight with tiger warrior vs dragon wraith warrior, both trading devastating blows at high speed, camera swinging low and fast as sparks explode on every clash, both warriors locked blade to blade face to face, then the dragon wraith breaks, jaw opening wide, a massive torrent of crimson fire erupts from his mouth directly into the tiger warrior's face, his head snapping back from the impact, massive body convulsing under the heat, clawed hand grabbing his own face as the fur scorches black and skin cracks beneath the fire, knees buckling under the force, sword arm dropping
Extreme wide shot — thousand cavalry on a fog-covered plain at dawn, camera positioned low in the grass, hooves thunder toward lens, slow motion as the first wave crests a hill and sunlight breaks behind them in blinding corona, cut to tracking shot riding alongside at full gallop, manes and armor catching light, crash zoom to general's raised sword, then snap to aerial crane pulling back to reveal the full impossible scale of the charge, earth trembling implied through handheld micro-vibration, Gladiator / Lawrence of Arabia grammar, 65mm film grain
Quality: Japanese full-color animation, high-speed cuts, numerous frames, 24fps, photorealistic, high-definition 2.5D. A 2.5x-speed action movie.
Cuts: Video created from a variety of camera angles, including overhead shots and drone footage.
Music: EDM
Scene 1:
An evil cyborg appears from a blind spot and is slashed at lightning speed with a Japanese sword.
Scene 2: With his back to the camera, he engages in a fierce battle with the cyborg, slashing with his sword at lightning speed.
The cyborg fires a laser beam, and the woman blocks the laser with her sword.
Scene 3: The woman slashes the cyborg at lightning speed, shredding it into scrap.
Scene 4: The resulting scrap explodes.
Full-body kung fu sword performer moving at extreme speed.
Action:
The character performs traditional kung fu sword forms using a single-edged Chinese-style or katana-inspired blade.
Movements include slashing arcs, thrusting motions, blade deflections, spins, overhead and low sweeps, aerial movements, stance transitions, rotations, jumps, and sudden controlled stops.
The body and sword move in perfect harmony, with every step and turn synchronized to precise blade control.
This is a cinematic martial arts performance, not realistic combat.
Motion Style:
Fluid yet powerful martial arts flow.
Smooth transitions between stances and techniques.
Dynamic rhythm with bursts of speed and moments of stillness, showing master-level control.
Elemental Effects:
Each sword movement is enhanced by changing elemental visual effects.
Elements include fire, lightning, wind, ice, glowing energy, and shockwave-like pulses.
Every motion briefly produces a different visual element — such as flaming trails, lightning streaks, wind pressure lines, ice patterns, or glowing energy flares.
Elements switch rapidly and unpredictably with each movement.
Visual Style:
Stylized cinematic martial arts.
Motion blur on sword trails, bright flashes on movement peaks, glowing arcs, and floating elemental particles.
Camera fixed in front, full body always visible.
Dark or neutral background to highlight motion, light, and visual effects.
"The camera orbits around Two samurai in traditional armor face each other in bamboo forest. Shafts of morning light cut through mist. At 1s, both explode forward simultaneously. 2s: katanas meet in center frame, steel rings out, impact sends visible shockwave through mist. 2.5-4s: rapid exchange of three strikes, blades catching light with each swing. 4.5s: they pass each other, freeze in opposing stances. Kurosawa framing, high contrast black and white with selective color on blood-red cord details."
Wild aerial anime fight against a giant monster, agile fighter spirals upward around the creature’s neck and shoulders, rapid slashes delivered at extreme speed, violent near-misses with grasping hands, exaggerated motion arcs, split-second impact freezes, cinematic camera orbiting the massive scale, brutal kinetic chaos
A cinematic, wide-angle slow-motion shot of a samurai made entirely of transparent, rushing crystal-clear water fighting a dragon made of thick, billowing black smoke. They are fighting on a wet rooftop in Tokyo at night.
Camera Settings: Shot on Arri Alexa 65, Panavision C-Series Anamorphic lens, 24mm wide angle, f/1.8 aperture for shallow depth of field, high shutter speed to capture crisp water droplets.
Lighting: High-contrast neon-noir lighting. Strong volumetric backlighting from pink and blue neon signs to maximize light refraction through the water samurai's body. Dramatic rim lighting to outline the black smoke dragon against the dark sky. Ray-traced reflections on the wet rooftop.
Key Details: 1. The neon city lights must refract and shine THROUGH the transparent water body. 2. The Smoke Dragon is volumetric and shifting. 3. Hyper-realistic fluid simulation, 8k resolution, epic scale.
A fierce Viking warrior with fur-lined armor and a massive engraved sword charges into battle in a snowy medieval village. The sky is dark and stormy as snow swirls violently through the air. Enemy raiders rush toward him from between the wooden houses. The Viking swings his sword with power, sparks flying as blades collide. His wolf companion snarls and leaps into the fight, circling and darting between attackers. Cinematic slow-motion moments capture the warrior blocking an axe strike, fur cloak whipping in the wind, and snow exploding from the ground with every clash. Add dynamic camera moves.
Scene: A traditional wooden dojo with sunlight streaming through rice paper windows. Dust motes dance in the light. Element: An old master with a long white beard and a young, eager student. Movement: The student attacks with a flurry of punches; the master blocks them all with one hand behind his back, barely moving his feet. Audio: [Object: Cloth Robes] sharply [Action: Snap] + [Sound Effect: Whipping Sound] + "Focus your breathing, not your fists." + Wise + Slow + Gentle + [Master speaking]. Other: Martial Arts / Fluid Motion / Steady Cam.
Scene: Cinematic action movie, matrix style.
Action:
The Coffee: The man and woman violently throw their coffee cups aside (out of frame). Brown liquid splashes in the air.
The Fight: Immediately, they turn to face each other aggressively. The woman throws a high kick towards the man, and the man raises his arm to block. They are engaging in fast martial arts combat.
Clothes: Their black suits move dynamically with the fight.
Audio: Intense fighting sounds. "Whoosh" (kick), "Smack" (block), glass breaking sound. Fast-paced techno combat music.
Camera: Side view tracking shot. The camera moves to the side to capture the profile of them fighting. Shaky cam for impact.
Action: The samurai bursts out of the paper, landing in the real world with a shockwave, then spins and slashes in a flurry of sword strikes. Camera: Quick orbit shot followed by dynamic tracking as he moves forward with each attack. Atmosphere: Ink splashes become 3D particles that fly off with each move, creating a surreal effect. Style: Fast-paced, energetic with traditional sumi-e style mixed into the environment.
intense martial arts duel on a rain-soaked urban street at night. Two fighters move with lightning speed, kicking up puddles and splashing rain. Neon signs reflect off wet pavement, and sparks fly from metal clashes. Camera locked at medium distance, capturing flips, strikes, and blocks in dramatic slow-motion bursts. Hyper-realistic textures, cinematic lighting, dynamic reflections, zero camera motion.
Low-angle tracking shot: A knight in battle-worn armor strides across a burning battlefield, sword dragging along the ground and sparking with each step. The camera pushes forward through smoke and falling ash, framing shattered catapults and silhouettes of clashing soldiers in the distance as the knight raises his blade to the sky. --ar 1:1 --motion high --video 1
Low-angle tracking shot: A knight in battle-worn armor strides across a burning battlefield, sword dragging along the ground and sparking with each step. The camera pushes forward through smoke and falling ash, framing shattered catapults and silhouettes of clashing soldiers in the distance as the knight raises his blade to the sky.
Tracking dolly shot: Two fighters brawl inside a skyscraper with shattered walls. The camera swings wide, capturing their silhouettes against a 200-story drop, shards of glass still falling into the city below. Every punch teeters dangerously close to the edge. Thrilling, cinematic, nerve-wracking.
The red warrior drops low into a crouch, spinning gracefully into a wide leg sweep. Dust lifts as his leg cuts across the ground in a powerful arc. In perfect timing, the man in white leaps high, body twisting as he clears the sweep with a sharp airborne motion. The camera tracks the sweep in a low dolly shot before tilting up to follow the leap. Wide framing shows the full choreography against the looming stone guardian, emphasizing the precision and control of both fighters. Fluid, elegant, martial artistry.
A lone samurai in a tattered black kimono walks slowly across a deserted battlefield at dusk, the ground littered with broken spears and burning banners; his sword drips with rain as thunder rumbles in the distance, and a sudden gust of wind sends ash swirling around him—camera tracks from behind in a steady dolly shot, then shifts to a close-up of his determined eyes under a straw hat, warm orange sunset light colliding with cold blue storm clouds, medium telephoto lens, cinematic contrast; the style is realistic historical drama with an epic, solemn mood.
Two samurai duel on an ancient scroll — every step they take unrolls more of the battle scene — the camera follows the unfurling paper until the fight bursts out into a full-color anime world
Create an 8-second animated sequence featuring 2 mysterious Shinobi ninjas sparing using ninjatō swords in a traditional Shinobi dojo. The dojo should have a minimalist design with wooden floors, tatami mats, and wooden walls, maintaining an open structure to allow for extensive movement and sparring activities. The ninja characters are dressed in traditional attire with a hood.
One ninja should be surrounded by swirling red energy trails that emanate from his body as he performs fluid, martial arts movements. The energy effects should pulse and flow, emphasizing their power and agility.
The second ninja should be surrounded by swirling blue energy trails that emanate from his body as he performs fluid, martial arts movements. The energy effects should pulse and flow, emphasizing their power and agility.
In the background, include traditional training weapons such as shuriken, kunai, and possibly a kyoketsu-shoge, mounted on the walls or placed around the dojo. Add scrolls and a few calligraphy banners with Japanese characters, reflecting themes of stealth, endurance, and martial prowess, hanging on the wooden walls. The overall color palette should be dominated by natural wood tones, with the glowing red energy providing a stark contrast. Subtle smoke effects at the base of the dojo can enhance the mystical atmosphere, while the open design of the dojo ensures the focus remains on the ninja's dynamic movements and the glowing energy trails.
{
"video": {
"duration": 8,
"fps": 30,
"resolution": "1920x1080",
"style": "cinematic, animated, martial arts film",
"setting": {
"environment": "traditional East Asian dojo with paper lanterns and wooden architecture",
"lighting": "dramatic, with soft ambient light and shadows",
"atmosphere": "tense, action-packed"
},
"characters": [
{
"type": "duck",
"appearance": "anthropomorphic, with vibrant feathers (one white with black accents, one brown with green accents), wearing traditional martial arts headbands",
"actions": [
"performs high kicks, spins, and punches in a stylized kung-fu manner",
"moves with agility and precision, feathers ruffling with each move"
]
},
{
"type": "duck",
"appearance": "anthropomorphic, with dark feathers and a red crest, wearing a different colored headband",
"actions": [
"blocks and counters with swift movements, uses wing strikes and beak jabs",
"leaps and dodges with exaggerated martial arts flair"
]
}
],
"sequence": [
{
"time": 0,
"description": "Two ducks face each other in a ready stance, surrounded by paper lanterns and wooden dojo elements, mist lightly rising from the ground."
},
{
"time": 1,
"description": "The white duck initiates the fight with a spinning kick, feathers flying as it moves."
},
{
"time": 2,
"description": "The brown duck blocks the kick and counters with a rapid series of punches, creating a blur of motion."
},
{
"time": 3,
"description": "Both ducks leap into the air, performing simultaneous high kicks that almost connect, with dramatic shadows cast on the dojo walls."
},
{
"time": 4,
"description": "The white duck uses a wing strike, sending the brown duck stumbling back slightly, feathers scattering."
},
{
"time": 5,
"description": "The brown duck recovers and charges forward with a series of beak jabs, forcing the white duck to dodge and weave."
},
{
"time": 6,
"description": "Both ducks pause momentarily, circling each other with intense focus, their breathing visible in the mist."
},
{
"time": 7,
"description": "The fight resumes with a flurry of synchronized movements, ending with both ducks in mid-air, about to clash again."
}
],
"audio": {
"background": "traditional East Asian instrumental music with a fast tempo, punctuated by martial arts sound effects (kicks, punches, impacts)"
},
"text_overlays": [
{
"time": 0,
"text": "Duck Kung-Fu Duel",
"position": "top-center",
"style": "bold, white with black outline, traditional font"
}
]
}
}
Intense close-up of a fierce Kung Fu duel — a striking, agile martial artist launches a powerful mid-air spinning kick toward the head of a towering opponent. Their expression is razor-focused, eyes locked, muscles coiled with explosive energy. The opponent, a disciplined Kung Fu master with weathered features and sweat-drenched robes, leans back in surprise, caught off guard. High contrast, dynamic motion blur, traditional Chinese setting, dramatic lighting, cinematic action shot.
A vertical American comic book page in the style of Simon Bisley, mixing divine iconography with brutal violence and anatomical exaggeration. A battle-scarred archangel in cracked golden armor plummets from the sky with a flaming spear, wings tattered and bleeding. Below, horned titans rise from storm clouds. One panel shows the angel roaring as he impales a beast through the mouth. Palette of radiant gold, ash white, and blood red. Glorious, apocalyptic, and sacredly violent. Dialogue and narration in illuminated manuscript-inspired English comic fonts. No real-world likenesses. --ar 9:19 --motion high --raw --video 1
Anime-style combat sequence: two rivals fight in an abandoned factory — the camera bounces with each impact, handheld-style — steel sparks, crates shatter, and their final clash sends a shockwave through the walls
Anime-style fight scene: two characters exchange blows in a tight alleyway — the camera moves side-to-side in sync with each parry, every strike lighting up the walls with glowing energy trails
[cinematic wide-angle footage], [low-angle perspective capturing the chaos of battle], [Roman soldiers charging through a snowy, forested battlefield], [gleaming armour with streaks of dirt], [background of smoke and fire against a muted gray sky], [golden tones from torches contrasting with dark shadows], [dynamic motion blur adding energy to the action], [shot on Kodak Vision3 250D for epic cinematic realism].
A cinematic fast cut rooftop fight scene at dusk. Opens on a wide aerial shot circling two fighters. Cut to a low-angle close-up of feet pivoting on gravel. Snap to handheld over-the-shoulder punch impact. Cut to extreme macro on blood dripping onto concrete. Final shot: wide silhouette as one figure stands tall while the other falls. VO (male, gritty tone): "Some battles are silent."
"A Star Ranger engages in a vibrant battle against a deadly red spider robot, depicted in dynamic anime style. The scene unfolds under a dark, starry nighttime sky, illuminated by bursts of energy from their clash. Background movement highlights fluttering debris and sparks, while rich reds and glowing blues create a visually stunning contrast. Rapid motion emphasizes the intensity as the Ranger evades the robot's attacks with acrobatic grace."
Two elemental warriors clash in an open field, one wielding glowing water and the other molten lava. The camera moves dynamically around the battle as their attacks create explosions of light and energy
将视频 1 向前延长 15s。角色 A 为穿黑色燕尾服、白色衬衫、黑色领结及米色马甲的深棕色短发男性，角色 B 为穿米色泡泡袖复古长裙、裙身带黑色蕾丝花边、佩戴深色水滴耳环的棕色盘发女性。舞厅内水晶吊灯光芒映照大理石地面，宾客们三两交谈持杯低语，一段悠扬的华尔兹弦乐前奏缓缓响起，人群的谈话声逐渐安静。角色 B 从舞厅一侧缓步走来，裙摆轻拂过大理石地面，目光环顾四周。角色 A 从人群中走出，穿过宾客径直来到角色 B 面前，微微欠身伸出右手，目光温柔注视着她。角色 B 垂眸浅笑，将右手轻轻搭上角色 A 的掌心，两人相视一笑，角色 A 引领角色 B 步入舞池中央，周围宾客自然后退让出空间，目光齐聚二人。角色 A 左手握住角色 B 右手举起，右手轻扶她的腰后，两人摆好华尔兹舞姿，随弦乐正拍踏出第一步，开始旋转。19 世纪宫廷舞会的典雅氛围，暖金色调。
FORMAT
Duration: 10 seconds
Aspect Ratio: 9:16 vertical
Style: cinematic photorealism, high-end fantasy beauty film, realistic skin and water physics, elegant and delicate
Camera: intimate close-up / medium close-up, subtle handheld movement, shallow depth of field
CHARACTER
An adult mermaid princess in her mid-20s, breathtakingly beautiful, ethereal and refined.
Long wet platinum-silver hair, pale luminous skin covered with tiny water droplets, large blue-green eyes, delicate pearl and shell accessories, subtle iridescent scales around her shoulders and collarbone.
CRITICAL ANATOMY RULE:
She is a real mermaid.
From the waist downward, her body transitions continuously into ONE complete iridescent fish tail with a large translucent tail fin.
NO human legs, NO thighs, NO knees, NO feet, NO transformation into legs at any point.
The tail remains clearly visible throughout the video.
She wears delicate pearl jewelry and translucent sea-inspired fabric around the upper body, elegant and slightly sensual but refined.
SETTING
Nighttime rocky shoreline beneath soft moonlight.
She rests half-submerged in shallow seawater beside dark wet rocks.
The ocean behind her reflects silver-blue moonlight.
Small waves move naturally around her tail and sparkling water droplets cling to her skin and hair.
0–3s
Medium close-up.
The mermaid princess is resting against a wet rock, her long shimmering tail curling naturally behind her in the water.
She notices the camera.
She slowly lifts her gaze toward the viewer with genuine curiosity, blinking once.
Her expression is innocent, thoughtful, almost as if she is studying a strange human creature.
Subtle ocean breeze moves strands of her wet hair.
3–8s
The camera slowly pushes closer.
She tilts her head slightly and looks directly into the lens.
In a soft, naturally curious female voice, she asks:
“So… humans really wear shoes every day?”
Natural realistic English lip sync.
The delivery should sound genuinely puzzled rather than seductive or theatrical.
After saying “shoes,” her eyebrows lift slightly, as though she still cannot understand the idea.
8–10s
She pauses.
Her eyes briefly glance downward, imagining human feet and shoes, then return to the camera.
A tiny amused smile appears.
Her tail fin gently flicks through the water behind her, creating a soft splash.
Hold eye contact for the final moment.
VISUAL STYLE
ultra-detailed cinematic fantasy realism
soft silver-blue moonlight
pearlescent highlights
wet skin texture
realistic water reflections
delicate facial micro-expressions
natural blinking and breathing
high-end fantasy film cinematography
85mm portrait lens feeling
shallow depth of field
soft background bokeh
romantic ocean atmosphere
AUDIO
Natural ocean waves
soft water movement
subtle nighttime sea ambience
clear intimate female voice
No background music
AVOID
human legs
human feet
split tail
two tails
tail transforming into legs
extra limbs
anime style
cartoon appearance
overacting
exaggerated seductive expressions
fast camera movement
large subtitles
text on screen
15-second cinematic photorealistic sci-fi action film, 16:9.
Earth is completely dry and abandoned.
A gigantic spacecraft carries the LAST OCEAN inside a transparent spherical containment field.
The enormous blue ocean floats inside the ship like a miniature planet.
A hostile fleet suddenly attacks.
Human defense ships race around the ocean carrier, protecting it.
Enemy fighters fire missiles.
The carrier takes a direct hit.
The containment field cracks.
A colossal wave of water bursts into space.
Human fighters immediately fly around the escaping water, using their ships to protect the floating ocean from enemy weapons.
The water forms a gigantic blue sphere against the stars.
The carrier's containment system activates again and captures the ocean just before it is lost.
The fleet accelerates toward a distant green planet.
Epic emotional science fiction, crystal-clear premise, spectacular water physics in space, intense fighter combat, huge scale, dynamic camera movement, photorealistic spacecraft, dramatic lighting, no text, no logos, no anime, no cartoon.
Prompt video 2 :
SCENE CONTEXT
Create a 20-second, 1080p ultra-realistic personal home video outside a small convenience store in Seoul. A hungry young Korean woman holding partially eaten bread breaks down while speaking to the young man beside her. His quiet reassurance makes her finally reach for him and cry in his arms. Emotional, intimate, and physically natural.
ACTIVE REFERENCES
Use the attached image as the exact opening frame and visual reference for both faces, hairstyles, clothing, seating positions, bread, wrapper, water bottle, shopping bag, bench, brick wall, and store entrance. The image controls visible details and initial object placement.
CHARACTERS
Woman: Korean, 23, beautiful natural face, tired reddened eyes, damp cheeks, loosely tied black hair with strands falling around her face. Worn beige cardigan, gray shirt, faded blue jeans, scuffed white sneakers. Her attempt to remain composed gradually gives way to trembling lips, broken speech, and genuine sobbing.
Man: Korean, 26, short black hair, navy windbreaker over a white T-shirt, dark trousers, gray sneakers. He listens attentively, responding with a serious, compassionate expression.
An off-screen friend holds the camcorder.
ENVIRONMENT
Weathered wooden bench against a red-brick wall beside an open convenience-store entrance. Woman seated on the left, man on the right. Begin with the bottle upright between them and the shopping bag behind it, matching the image. Cool overcast afternoon, soft daylight, fluorescent store lighting visible inside.
CAMERA & TEXTURE
Raw early-2000s consumer DV footage filmed by a friend: strong natural handheld shake, spontaneous imperfect framing, occasional autofocus hunting, exposure shifts, motion blur, faded colors, soft digital detail, mild noise, and brief accidental zooms. The camera follows the activity with natural amateur movements.
The operator stands in front of the bench, slightly toward its right end. Begin at the reference framing. A hesitant short zoom follows the woman’s trembling expression. As she leans toward the man, the operator reacts slightly late, then takes half a step backward to keep their embrace visible. Preserve readable faces and gestures.
FORMAT MODE
Four timed segments with exact hard cuts at 5, 10, and 14 seconds. End at exactly 20 seconds. Keep actions and object positions continuous across cuts.
ACTION & TIMING
0–5 seconds:
The woman finishes swallowing, still holding the partially wrapped bread with both hands. She slowly raises her wet eyes toward the man. Her lower lip trembles.
In a strained Korean voice, she says, “저… 너무 힘들어요.”
She falters after the first word, trying to control her breathing. The camera makes a small, imperfect zoom toward her face.
5–10 seconds:
Tears run down her cheeks as she adds, “아무도 제 곁에 없어요.”
She lowers the bread and places it carefully on its opened wrapper on the bench to her left, freeing both hands. While listening, the man lifts the sealed bottle with his right hand and places it upright on the pavement beside his right shoe. He then moves the shopping bag from between them to the bench on his right, clearing the space.
10–14 seconds:
The man turns his torso toward her and quietly says, “제가 옆에 있을게요.”
He maintains gentle eye contact. She looks at him, draws a shaky breath, and her expression crumples. She turns toward him and begins reaching with both arms. The camera pans slightly late to follow her movement.
14–20 seconds:
She shifts closer along the bench and wraps both arms around his torso, pressing her face against his shoulder. He briefly pauses in surprise, then gently returns the embrace, one hand resting on her upper back and the other supporting her far shoulder.
Her shoulders tremble with uneven sobs; her fingers clutch the back of his windbreaker. He softly says, “울어도 괜찮아요.”
The operator takes half a step backward, causing a brief footstep shake, then holds on their seated embrace. End with her still crying against his shoulder while he quietly holds her.
AUDIO
Natural Korean dialogue exactly as written, with accurate lip-sync and distinct voices. Her speech becomes breathy and broken before turning into audible sobbing. His voice remains low and reassuring. Include wrapper rustling, jacket fabric shifting during the hug, distant traffic, faint refrigerator hum, and subtle camcorder-handling noise. No background music or narration.
CONTINUITY
Preserve the reference identities, clothing, natural skin texture, bench geography, and lighting. The bread remains partially eaten and rests on its wrapper to the woman’s left before the embrace. Move the bottle and shopping bag visibly before she shifts closer; they remain in their new positions. Both characters stay seated. Her hands are empty during the hug. Maintain realistic arm placement, body contact, accumulating tears, and natural breathing without exaggerated theatrical movements. Fully photorealistic live action. No slow motion, beauty filters, stylized lighting, subtitles, added text, logos, or watermark.
Prompt video 1 :
SCENE CONTEXT
Create a 30-second, 1080p ultra-realistic personal home video capturing a quiet, heartbreaking encounter in an older Seoul neighborhood. A beautiful young Korean woman, exhausted and hungry, sits outside a small convenience store. A young Korean man notices her and shares his food. Her restrained reaction carries the emotion: embarrassment, disbelief, then relief at being treated kindly.
ACTIVE REFERENCES
Text-only generation. The written descriptions define the consistent characters, objects, and locations throughout.
CHARACTERS
Woman: Korean, 23, naturally beautiful, pale tired face, slightly chapped lips, long black hair loosely tied back with stray strands around her cheeks. Worn beige cardigan over a plain gray shirt, faded blue jeans, scuffed white sneakers. Natural skin texture, no glamorous makeup. She moves slowly and avoids eye contact, trying to hide her hunger.
Man: Korean, 26, short black hair, gentle expression, navy windbreaker, white T-shirt, dark trousers, gray sneakers. His concern is expressed through patient, ordinary gestures.
His off-screen friend holds the camcorder, initially recording their walk.
ENVIRONMENT
An older Seoul residential alley on a cool, overcast afternoon. A small convenience store faces the pavement; a weathered wooden bench stands beside its entrance, with a red-brick wall behind it. The woman sits at the left end. The man approaches from the right carrying one small transparent shopping bag containing one wrapped bread roll and one sealed water bottle. Soft daylight, distant neighborhood traffic, a faint refrigerator hum from the open store entrance.
CAMERA & TEXTURE
Raw early-2000s consumer DV footage filmed by a friend: strong natural handheld shake, spontaneous imperfect framing, occasional autofocus hunting, exposure shifts, motion blur, faded colors, soft digital detail, mild noise, and brief accidental zooms. The camera follows the activity with natural amateur movements.
The friend begins one step behind the man, then stops near the bench’s right end, leaving the woman space. Walking creates visible footstep shake. When the man stops unexpectedly, the camera briefly drifts past him before correcting. Later details are captured with one hesitant short zoom. Keep the interaction readable and quietly observed.
FORMAT MODE
Five timed home-video segments with exact hard cuts at 6, 12, 18, and 24 seconds. End at exactly 30 seconds. Preserve the same bench, character positions, clothing, food, and bag across cuts.
ACTION & TIMING
0–6 seconds:
The camera follows behind the man approaching the store. On the bench, the woman searches an empty cardigan pocket, then lowers her hand into her lap. She notices the bread through his shopping bag; her gaze lingers before she quickly looks down, embarrassed. He notices and stops. The operator takes one extra step, then corrects the framing.
6–12 seconds:
From beside the bench, a slightly crooked medium shot. The man bends his knees to meet her eye level without crowding her. Holding the bag in his left hand, he retrieves the wrapped bread with his right and gently offers it.
He quietly says in Korean, “이거 드세요.”
She looks from the bread to his face, hesitating, then accepts it with both hands. He releases it only once her grip is secure.
12–18 seconds:
The man sits at the bench’s right end, leaving a respectful gap. He places the bag on the bench to his right, removes the sealed water bottle, and sets it upright between them.
The woman rests the bread on her lap and peels back the wrapper with trembling fingers. The camera makes a small imperfect zoom toward her hands; autofocus briefly searches, then settles. She lifts the partially wrapped bread and takes one small bite.
18–24 seconds:
A closer, imperfectly framed view of her face. She chews slowly and swallows. Her shoulders loosen, but her eyes fill with tears. She pauses with the bread held over her lap, presses her lips together, and wipes one cheek with her free hand.
In a barely steady voice, she says, “감사합니다.”
The camera holds quietly, with subtle hand tremor.
24–30 seconds:
A wider view includes both of them. He gives a small reassuring nod and says softly, “천천히 드세요.”
He then looks toward the alley, giving her privacy. She takes another small bite, holding the wrapper beneath the bread to catch crumbs. Her gaze drops, tears still visible, but her breathing settles. End on both sitting quietly together, the untouched water bottle standing between them.
AUDIO
Natural Korean dialogue only as specified. Distant passing vehicles, faint footsteps, refrigerator hum, bread-wrapper rustling, quiet breathing, and subtle camcorder-handling noise. Her voice is fragile and restrained; his voice is gentle and conversational. No background music or narration.
CONTINUITY
Maintain consistent faces, ages, hairstyles, clothing, and footwear. One bread roll transfers from the man’s bag to the woman and remains partially wrapped; each bite visibly reduces it. One water bottle remains sealed and upright between them. The shopping bag stays at the man’s right after being placed there. Preserve realistic hands, chewing, swallowing, restrained tears, and the bench’s geography. Use ordinary overcast daylight throughout. Fully photorealistic live action; no stylized lighting, slow motion, beauty filters, subtitles, added text, logos, or watermark.
Photoreal cinematic short, 30 seconds, one unbroken camera move, 24–30fps, natural indoor lighting mixed with TV glow. No cuts, no gore, no blood, no logos. No burned-in subtitles. All lines are spoken, not written on screen.
Setting: a small lived-in apartment at night. Couch facing a large TV. Two controllers. Soft lamp. Half-finished snacks. A second headset on the cushion. On the TV: a bright third-person co-op fantasy game, split or shared screen, two adventurers running together through glowing ruins.
Characters in the room:
Player 1 — young man, hoodie, focused then laughing, left side of the couch.
Player 2 — young woman on his right, same age, casual clothes, laughing with him, second controller, leaning into his shoulder when they win.
00:00–00:08
Start on the TV: both in-game characters sprint and tag-team a small monster. Camera slowly pulls back from the screen to the couch. The couple cheers. He shouts at a jump. She bumps his arm, grinning, mashing buttons. While they play he talks to her, excited: “We’re so good together. Look at that combo. Nobody plays like us.” She laughs and keeps playing. They look at each other, not just the screen. Warm co-op night. On the TV her avatar stays fully present the whole time.
00:08–00:16
They clear a checkpoint. In-game victory chime. Her character is still fully in the game, running beside his. He keeps talking to her: “That was perfect. You’re carrying us. We still got it.” She raises the controller like a trophy. He laughs. They turn toward each other for a high-five / clap.
00:16–00:24
This is the transition. He swings his hand in for the clap. In the first part of the motion she is still there, smiling, hand coming up to meet his. Mid-clap she fades from the couch: her hand first, then her body, until his palm hits empty air. No one is on the right seat. One body print. Second controller rests unused, battery light dark. Headset unused.
Rule for the game screen: Player 2’s avatar remains in the game for as long as the girl is still visible on the couch. The instant she is fully gone from the room, her adventurer fades out of the game and the slot goes empty. Not before.
00:24–00:30
His smile stalls. He stares at the empty space where her hand should have been. Quietly, broken, he says: “I wish you are still here.” The camera drifts to a small framed photo of the girl on the side table. A single candle burns in front of it, warm light on her face. He sets his controller down. Hold on the candlelit photo, empty seat behind it, co-op screen now showing only Player 1. No cause of death. No grave. No hospital.
Audio: co-op music, his talk about how well they play, her laugh, two controllers, then music thins after the missed clap, her voice gone, one controller, fridge hum, distant traffic, leftover victory chime, candle flicker, his last line.
Do not include: gore, wounds, funeral, jump-scare ghost face, horror makeup, first-person death footage, guns, on-screen logos.
Unplanned handheld 8mm home-movie recording, shot by an older teenage brother moving with the group, imperfect framing, warm faded film colors, visible fine grain, slight gate weave, occasional motion blur, natural summer daylight, realistic deep focus.
Characters:
The hero is a thirteen-year-old Japanese boy with slightly overgrown black hair, wearing a faded white baseball shirt with red piping, navy shorts, worn canvas sneakers and a red cloth shoulder bag.
His three friends: a stocky boy wearing a yellow baseball cap and striped T-shirt, a thin boy with round glasses wearing a pale-blue short-sleeved shirt, and a smaller energetic boy wearing a white sleeveless shirt and khaki shorts.
They carry insect nets, metal canteens, flashlights and a hand-drawn treasure map folded inside an old school notebook.
They tease one another, argue over directions and laugh loudly as they travel together.
Setting:
Rural Japan during the late Shōwa era, summer of 1978 — green rice fields, narrow country roads, wooden houses with tiled roofs, concrete utility poles, rusted road signs and distant forested mountains.
An abandoned local railway line runs beyond a weathered crossing, disappearing through tall grass into a dark railway tunnel.
Beyond the tunnel stands a forgotten wooden signal cabin beside a shallow river, surrounded by cicadas, weeds and golden late-afternoon sunlight.
No modern objects, no contemporary buildings, no smartphones, no modern cars, no LED signs.
Shot list:
0.0s–4.0s — Wide tracking shot, waist height, camera moving beside the group
The four boys race old Japanese bicycles along a narrow road between bright green rice fields, shouting and laughing as their shirts flutter in the summer wind; the handheld camera struggles to keep pace while cicadas roar and bicycle bells ring.
4.0s–7.0s — Medium-wide, eye level, sudden stop and loose reframe
They skid to a stop at a rusted abandoned railway crossing. The hero jumps off his bicycle, pulls the folded hand-drawn map from his shoulder bag and points toward the overgrown railway tracks while the others gather around him.
7.0s–10.0s — Rear tracking shot, low angle, uneven handheld movement
The boys leave their bicycles hidden in the tall grass and run along the disused tracks, balancing on the rails and pushing one another playfully; their insect nets and metal canteens bounce as the camera follows from behind.
10.0s–13.0s — Medium shot, entering darkness, natural exposure shift
They reach the mouth of an old railway tunnel and slow down. The hero switches on a flashlight, raises it above his head and leads the group inside as the bright summer landscape behind them rapidly fades into darkness.
13.0s–16.0s — Close handheld group shot inside the tunnel
The flashlight beam shakes across damp brick walls while the boys walk closely together, trying to act brave. A sudden flutter of birds erupts from the ceiling, making all four scream, duck and then burst into embarrassed laughter.
16.0s–19.0s — Wide shot, camera following them out of the tunnel
They run from the tunnel into brilliant golden sunlight and stop in surprise. Below them lies a sparkling river, an old wooden signal cabin and a forgotten railway platform almost completely covered by summer grass.
19.0s–22.0s — Medium tracking shot, running downhill
The hero checks the map, spots the signal cabin and races toward it. His friends chase him through waist-high grass while the camera follows with excited unstable framing and brief flashes of sunlight across the lens.
22.0s–25.0s — Close medium, crouched handheld angle
Inside the dusty signal cabin, the hero kneels beside a loose wooden floorboard and pulls out a rusted metal lunchbox. The four boys crowd around as he opens it, revealing old glass marbles, a brass compass and a faded black-and-white photograph.
25.0s–28.0s — Tight group shot, drifting between faces
The boys quietly examine the small objects. Their excitement becomes a shared look of wonder as the brass compass catches the warm sunlight; the smallest boy smiles and whispers something while the others nod.
28.0s–30.0s — Wide rear shot, camera settles
The four boys sit side by side on the abandoned platform with their backs toward the camera, looking across the river and distant mountains as the summer sun begins to set; the hero holds the brass compass in his hand while their bicycles remain far behind beyond the tunnel.
Audio:
Natural location sound accompanying the footage — loud cicadas, bicycle chains, rattling metal canteens, boys shouting and joking in casual Japanese, sneakers striking wooden sleepers, echoing footsteps inside the tunnel, startled screams, fluttering wings, flowing river water, distant train whistle and warm summer wind moving through the grass.
The Japanese conversation is spontaneous and overlapping, mostly indistinct except for short reactions and laughter.
No music.
Style:
Raw amateur late-1970s Japanese 8mm home-movie footage, nostalgic but realistic, warm sun-faded colors, fine analog grain, mild flicker, occasional overexposure, imperfect focus, natural handheld movement and authentic childhood spontaneity.
A grounded coming-of-age summer adventure about friendship, curiosity and the final days of childhood.
No polished commercial cinematography, no modern fashion, no fantasy elements, no supernatural effects, no subtitles, no on-screen text.
LONDON AFTER DARK | 15-Second Cinematic City Tour | WAN 3
Character & Visual Continuity
Begin with the exact supplied Soho couple reference image as the opening frame. Preserve both characters’ facial identities, hairstyles, natural skin texture, proportions, and outfits throughout. The man wears a fitted ivory henley, black jeans, and necklace. The woman wears a charcoal off-shoulder top, brown shorts, black tights, and carries a black handbag. Their chemistry is natural, romantic, and understated. Both protagonists remain visible in every scene.
Scene 1 — Soho, Wardour Street (0–2.5s)
Open on the supplied image. Both characters make direct eye contact with the camera for less than one second, then exchange a smile and begin walking together. His arm moves naturally from her waist as the camera tracks backward at close range. Neon bar signs, authentic London storefronts, and lively pedestrians fill the background.
Scene 2 — Carnaby Street (2.5–5s)
Hard cut to Carnaby Street beneath its illuminated overhead decorations. A lateral tracking shot follows the couple past busy shops and groups of friends. She playfully takes his hand and pulls him toward a storefront. He laughs and follows, maintaining natural movement and believable chemistry.
Scene 3 — Piccadilly Circus (5–7.5s)
Hard cut to Piccadilly Circus, with its recognizable illuminated advertising screens and red double-decker buses passing behind them. A dynamic close-range camera pass captures the couple pausing at a pedestrian crossing. She turns toward him and says something playful. He smiles as they continue across the street with surrounding pedestrians.
Scene 4 — Leicester Square (7.5–10s)
Hard cut to Leicester Square, surrounded by illuminated cinemas, restaurants, and active nightlife. A street performer entertains a small gathering nearby. The couple briefly turn toward the performance, exchange amused glances, and continue walking shoulder to shoulder. The camera tracks them from a medium-close angle, keeping the background active and authentic.
Scene 5 — South Bank, River Thames (10–12.5s)
Hard cut to the Thames riverside promenade, with the illuminated London Eye visible across the water. The couple walk hand in hand as the camera glides laterally beside them. He turns toward her, and she responds with a spontaneous smile. Reflections shimmer across the river while pedestrians move naturally through the background.
Scene 6 — Westminster Bridge (12.5–15s)
Hard cut to Westminster Bridge, with illuminated Big Ben and the Palace of Westminster clearly recognizable behind them. The couple stop close together. He gently wraps an arm around her waist, and she leans toward him. The camera makes a subtle curved movement into an intimate medium-close portrait. They exchange a brief glance before looking directly toward the camera. End on a confident, natural cinematic frame.
Cinematography & Direction
Six distinct scenes connected by clean, motivated hard cuts. Alternate backward tracking, lateral movement, close-range camera passes, and subtle curved camera movement. Keep both protagonists prominent in every shot. Avoid repetitive head-to-toe camera scans, static posing, excessive wide shots, and zooms.
Capture authentic London nightlife through recognizable architecture, British street signage, left-side traffic, red buses, active pedestrians, and believable street interactions. Every location should feel naturally populated without distracting from the couple.
Maintain consistent nighttime lighting with realistic neon illumination, warm storefront lighting, natural shadows, and accurate reflections. Preserve detailed skin texture, believable facial expressions, realistic walking mechanics, and consistent character identities.
Each new location appears through a clean cut, never through teleportation or morphing. No artificial slow motion, distorted hands, unnatural gestures, plastic skin, excessive smoothing, or animated appearance.
Audio: Natural London nightlife ambience, footsteps, passing traffic, background conversations, and brief spontaneous dialogue between the couple. No generated music.
Visual style: Hyper-realistic cinematic street photography, intimate editorial fashion-film aesthetics, dynamic camera movement, natural human interactions, and authentic London atmosphere. Total duration: 15 seconds.
ALMATY, KAZAKHSTAN — MORNING CITY TOUR V3
22 seconds, 16:9, WAN 3.
Six scenes, HARD CUTS ONLY. No flashes, fades or dissolves. Hyper-realistic morning in Almaty. Preserve the reference man's exact face, beard, brown beanie, tan jacket, gray hoodie, jeans and sneakers. Natural crowds, motion and light.
SCENE 1 (0–3s) — EXACT REFERENCE LOCATION
Begin on the supplied image. For ALL three seconds preserve the SAME concrete stairs, dark railings, concrete pillars, bushes, stickered walls, red circular sign with white writing, white numbered signs including 656, and sidewalk. Do not replace, rebuild or transform this background; no tree-lined street. He makes eye contact under one second, then walks naturally through this SAME location as camera tracks gently. Opening title appears ONLY here, two lines:
Almaty, Kazakhstan
Morning City Tour
At 3s hard cut; title vanishes instantly.
SCENE 2 (3–6s) — ASCENSION CATHEDRAL
Lateral medium tracking as he walks past the colorful wooden cathedral in Panfilov Park, briefly glancing toward it. Visitors behind him, natural sunshine. No text.
SCENE 3 (6–9s) — ZHIBEK ZHOLY CAFÉ
A barista hands him takeaway coffee. He accepts, smiles and says Kazakh: “Рақмет.” English subtitle: “Thank you.” Natural handoff and lip-sync. Camera tracks him leaving with the cup.
SCENE 4 (9–14s) — GREEN BAZAAR BUTCHER
Medium-close two-shot at a busy meat counter, showing both faces and beef cuts. He asks in Kazakh: “Бүгін не жақсы?” Subtitle: “What's good today?” Butcher maintains eye contact, gestures to beef ribs and replies in Kazakh: “Бүгін сиырдың қабырға еті жақсы.” Subtitle: “The beef ribs are excellent today.” Let subtitle remain for a brief reading beat. The man STAYS at the counter, smiles and nods in understanding while the butcher remains engaged. Neither turns or walks away. Then HARD CUT. No fruit.
SCENE 5 (14–18s) — REPUBLIC SQUARE
Low-angle tracking as he walks with the Independence Monument behind him, adjusting his beanie while holding coffee. Pedestrians move naturally. No text.
SCENE 6 (18–22s) — CENTRAL ALMATY CAFÉ STREET
Busy pedestrian sidewalk with outdoor café seating, storefronts and locals. Angled front-side medium view. He walks LEFT to RIGHT on the sidewalk, happy and relaxed; sips coffee, lowers cup, smiles, then fully EXITS frame right. Hold briefly and end with clean cut. No roadway walking, no final title. Storefront text must be realistic, stable and legible: use a simple red sign with credible Kazakh lettering, a green-and-white café or shop sign with correctly spelled words, and replace the nonsensical “HHRIHM” sign with “CAFE” in the same white lettering and style. No random symbols or invented brands.
SUBTITLES — ONE DESIGN THROUGHOUT
Poppins SemiBold 600, white #FFFFFF on softly rounded charcoal-slate #17212B at 60% opacity. Background translucent; consistent padding, bottom-center position and safe margins. Opening title two lines, first larger. Translate both café and butcher dialogue into American English. One subtitle at a time, synchronized to speech; hold butcher's answer briefly. No text outside specified title, subtitles and realistic storefront signs.
CONTINUITY / AUDIO
Same man and wardrobe throughout; coffee first appears in Scene 3 and stays consistent. Natural gestures, physical handoffs, accurate Kazakh lip-sync, realistic skin, fabric and architecture. No morphing, fake skin, warped signs, object teleportation, zooms, slow motion or empty shots. City, café and market ambience only. No music or narration.
Format: 9:16 vertical, 15 seconds, cinematic, ultra-realistic, massive subjects, high-detail, no text, no logos.
0–3 sec: Nighttime futuristic city street after heavy rain. Camera starts extremely low at street level, reflections of skyscrapers stretching across the wet pavement. Everything feels enormous. Suddenly, the entire street subtly moves as if the city itself is breathing—pavement rises and falls by a few centimeters. Cinematic fog drifting between buildings, realistic rain droplets, deep atmospheric sound.
3–6 sec: Camera rapidly pushes forward between towering buildings. Windows across hundreds of floors flicker simultaneously in a wave, travelling upward like a heartbeat. The camera tilts toward the sky as clouds begin rotating unnaturally above the city.
6–10 sec: A gigantic glowing geometric structure slowly emerges from behind the skyscrapers—not a creature, not a spaceship, something completely unfamiliar. It is impossibly tall, dwarfing the buildings. Its surface looks like black glass with thin luminous cracks flowing across it. Camera keeps pulling back to reveal its enormous scale.
10–13 sec: The structure suddenly opens vertically like a colossal doorway. A blinding white-blue light floods the entire street. Cars, rain and loose debris react to the powerful pressure wave. People appear only as tiny silhouettes far below, emphasizing scale.
13–15 sec: Smash cut to an extreme close-up of one tiny raindrop falling in slow motion. Inside the droplet, the entire gigantic city and mysterious structure are reflected perfectly. The droplet hits the pavement—CUT TO BLACK.
Visual style: premium theatrical science-fiction cinematography, physically realistic lighting, volumetric fog, wet reflective surfaces, dramatic scale, realistic physics, subtle handheld camera movement, anamorphic cinematic depth, high dynamic range, extremely detailed environments, natural motion blur, realistic reflections, powerful contrast, immersive atmosphere.
Camera: aggressive cinematic push-ins, low-angle scale shots, rapid perspective transition, controlled handheld movement, final macro shot.
Sound design: distant thunder, rain hitting pavement, deep sub-bass rumble, building vibration, rising atmospheric tone, enormous pressure-wave impact, sudden silence on the final cut.
Important: Keep the main subject large and dominant inside the vertical frame, avoid tiny distant compositions. No recognizable existing movie designs, characters, franchises, logos, text or copyrighted elements. Original concept and original visual design.
WAN 3.0 PRIME / 30 seconds / 16:9 / 1080p
Use @Image 1 as the character reference.
REFERENCE LOCK:
Preserve the exact overall look and identity inspired by @Image 1:
adult woman in her early 20s, Korean-idol-style beauty, long straight black hair with soft natural volume, fair porcelain-like skin, large expressive eyes, delicate V-line face, small nose, naturally glossy lips, light natural makeup, approachable and cute expression.
OUTFIT LOCK:
Match the outfit to @Image 1 and keep it consistent through the whole video:
white hoodie layered under a dark oversized jacket, short black pleated skirt, white socks, white sneakers.
This is casual Japanese street fashion, relaxed and natural, not stage costume.
VISUAL STYLE:
A realistic Tokyo travel snapshot video shot on a smartphone.
26mm smartphone lens look, deep depth of field, realistic skin texture, mild HDR, neutral iPhone-like color, unedited snapshot feeling.
No cinematic color grading, no studio lighting, no dramatic rim light, no light streaks, no fake bokeh overload, no heavy motion blur.
Everything should feel like real casual footage shot during a trip in Japan.
PERFORMANCE STYLE:
Natural expressions, small gestures, casual real-person behavior, selfie-like energy in some shots.
She should feel friendly, cute, relaxed, and spontaneous.
EDITING STYLE:
Fast but readable travel MV rhythm.
Hard cuts, flash cuts, and natural movement-based cuts.
Keep the pacing energetic but grounded in realism.
AUDIO / DIALOGUE FEEL:
Heavy rhythmic beat music underneath.
Natural environmental sound can briefly be felt in some scenes.
Japanese dialogue lines should feel spontaneous and casual.
30-SECOND STRUCTURE:
0:00–0:02
Tokyo Tower selfie-style walking shot, low angle.
She walks while holding the camera selfie-style, Tokyo Tower clearly visible behind her.
The sky is slightly overexposed white.
Natural smile, relaxed eye contact.
Neutral daylight, deep depth of field.
Dialogue: 「とうきょうに きちゃった…！」
Hard cut on the beat.
0:02–0:04
Shibuya scramble crossing at night.
She steps into the crowd, then turns back toward the camera.
Neon signs remain sharp and readable in the background.
Realistic night exposure, no fake glow effects.
Fast rhythmic cut.
0:04–0:06
Convenience store interior.
She holds an onigiri toward the camera and raises one eyebrow slightly.
Cool fluorescent lighting, realistic skin texture, casual playful expression.
Rhythmic flash cut.
0:06–0:08
Yoyogi Park.
She sits casually and looks away from the camera.
Sunlight through trees creates dappled shadow.
A little lens dirt or light smudge can be visible on the image.
Wind and birds are subtly felt.
Cut back naturally with body movement.
0:08–0:11
Takoyaki food stall.
She takes a bite of takoyaki, reacts immediately, opens her eyes wide, then laughs.
Warm lantern light, amber tone, real skin texture.
Dialogue: 「あつっ！ めっちゃ おいしい！」
Cut right after the laugh lands.
0:11–0:13
Night vending machine shot.
She stands in front of a glowing vending machine.
Mixed LED light reflects naturally on her face.
She gives a mischievous little smile.
Single clean shot, no exaggerated lighting effects.
0:13–0:15
Train window shot.
Her side profile appears in the reflection on the train window.
She looks outside instead of at the camera.
The city outside is naturally soft, not heavily blurred.
Quiet reflective mood.
Flash cut.
0:15–0:17
Torii gate, low angle.
She walks toward the torii and passes beneath it.
The torii fills the upper frame.
Natural daylight, wide smartphone perspective.
Flash cut.
0:17–0:21
Digital light show, iconic shot.
She slowly raises her face.
Blue, purple, and gold projected light moves softly across her face.
Use gentle focus shift: her face stays clear while the light remains soft and diffused.
Only soft shadow and color play on the skin, no harsh reflective glare.
This part can feel slightly slower and more iconic, around 60% speed feeling.
Soft cut during her head movement.
0:21–0:23
Gundam statue reveal.
Start on her face in medium close-up, then pan upward to reveal the full Gundam.
She gives a small impressed smile.
Natural daylight, wide angle.
Sharp cut immediately after the smile.
0:23–0:25
Lantern alley follow shot.
Tracking shot from behind and slightly beside her as she walks through a narrow alley lined with lanterns.
Warm lantern color and natural shadow.
She turns around to look back at the camera.
Cut while moving forward.
0:25–0:28
Rooftop skyline at night.
Selfie close-up first, then she extends her arm to reveal the skyline behind her.
Wind moves her hair naturally.
Real smartphone night-city look, not overly cinematic.
Dialogue: 「この やけい、はじめて みた」
Cut sharply at the end of the line.
0:28–0:30
Nara Park.
She crouches and offers deer crackers.
A deer bows, she laughs in surprise, then another deer nudges her from behind so she almost loses balance.
Slight handheld shake is okay here.
Everything should feel like real captured footage.
Dialogue: 「ちょっと！ ちゃんと ばいばい したでしょ！」
End with a fade to black.
IMPORTANT:
Keep the girl visually consistent in all scenes.
She must always feel like the same person from @Image 1.
Do not change her hairstyle, age, or outfit identity.
Keep her attractive but natural, like a real travel snapshot subject.
NEGATIVE:
anime look, over-cinematic lighting, studio lighting, glossy beauty ad look, fashion editorial look, heavy color grading, light streaks, strong motion blur, unrealistic skin smoothing, fake HDR halos, unstable identity, face drift, hairstyle change, costume change, idol stage costume, unrealistic poses, extra fingers, distorted hands, deformed face, duplicate person, text, subtitles, logos, watermark, UI.
A futuristic experimental racing vehicle attempts to become the first ground vehicle to maintain Mach 1 across open terrain. Dawn. Engines ignite. The camera follows close behind as the vehicle accelerates across a vast desert. 200 mph. 400 mph. 600 mph. The landscape becomes a blur. Shockwaves begin forming around the body. Warning messages appear throughout the cockpit. The vehicle starts lifting slightly from the ground. Control surfaces struggle to stabilize it. The camera alternates between wide aerial shots and terrifyingly close wheel-level views. 700 mph. 750 mph. Small components begin tearing away from the chassis. The driver refuses to slow down. Mountains ahead approach far too quickly. The speed reaches the critical threshold. A massive sonic boom erupts across the desert. The vehicle becomes unstable. It drifts sideways at incredible speed. The driver fights for control. For several seconds disaster seems inevitable. Then the vehicle stabilizes and continues accelerating. Final shot: the machine races beyond the horizon while the sonic shockwave rolls across the desert behind it. Ultra-realistic cinematic speed record attempt, realistic aerodynamics, overwhelming speed sensation, shockwaves, intense tension, detailed cockpit, IMAX quality, dramatic scale.
Seoul Summer Day - Video Prompt
Create a 30-second, 1080p ultra-realistic documentary-style personal home video showing an ordinary summer day in the life of a young Korean man. The footage should feel spontaneous, intimate, imperfect, and genuinely observed rather than performed.
MAIN SUBJECT
The same young Korean man in his early 20s throughout the entire video.
Naturally handsome, realistic skin texture, minimal styling, relaxed expression, slightly tired but peaceful eyes.
He has naturally messy medium-length dark hair with a few strands falling over his forehead and very subtle stubble.
He wears a loose washed-black T-shirt, relaxed olive-beige trousers, worn white sneakers, and a simple silver wristwatch.
Keep his face, identity, body proportions, hairstyle, clothing, watch, and overall appearance completely consistent from beginning to end.
LOCATION
A quiet older residential neighborhood in Seoul during a warm summer afternoon.
Narrow concrete alleys, low-rise homes, rooftop terraces, external staircases, potted plants, laundry lines, parked bicycles, utility poles, overhead wires, mature trees, concrete walls, small residential courtyards, and distant city sounds.
The neighborhood should feel authentic and lived-in.
No crowds, tourist attractions, advertisements, recognizable brands, or commercial activity.
CAMERA / VISUAL STYLE
Authentic casual personal-video footage captured with an older consumer digital camera.
Handheld camera operated by a friend walking nearby.
Natural camera shake, imperfect framing, occasional autofocus changes, slight exposure adjustments when moving between sunlight and shade, soft image detail, mild motion blur, subtle digital noise, slightly muted colors, imperfect white balance, and natural compression.
The camera operator occasionally reacts a little late, cuts off part of the subject, or briefly loses focus.
No stabilization, gimbal movement, drone shots, cinematic camera choreography, dramatic lighting, slow motion, modern commercial color grading, or polished cinematography.
The footage should feel like someone simply decided to record his friend during an ordinary day.
00:00–00:05 — MORNING ROOFTOP
He sits casually on a small rooftop beside an old plastic chair.
A cold bottled drink rests beside him.
He looks quietly across the neighborhood while the wind moves his hair and T-shirt.
He takes a sip, notices something happening in the distance, and smiles faintly.
He briefly notices the camera and gives a subtle nod before looking away.
The camera takes a moment to find focus on his face.
00:05–00:10 — WALKING THROUGH THE ALLEY
He gets up and walks downstairs into the neighborhood.
He walks casually through a narrow concrete alley with his hands in his pockets.
He passes parked bicycles, potted plants, laundry hanging from balconies, and old residential walls.
The camera follows several steps behind him.
He occasionally looks back toward the camera but never deliberately poses.
His footsteps remain naturally synchronized with his movement.
00:10–00:14 — SMALL EVERYDAY MOMENT
He notices an old basketball resting near a wall.
He picks it up, casually bounces it twice, then takes a simple shot toward a nearby hoop.
The shot misses.
He laughs quietly, shakes his head, and leaves the ball where he found it.
The camera briefly loses focus during the movement and recovers naturally.
No exaggerated athletic movement.
00:14–00:19 — LOCAL SHOP
He walks to a tiny neighborhood shop and buys a cold drink.
He exchanges a few natural words with the shopkeeper but the conversation is not clearly audible.
He steps outside, opens the bottle, takes a drink, and leans casually against the wall.
He watches bicycles and pedestrians passing in the distance.
The camera remains handheld and slightly imperfect.
00:19–00:23 — SUMMER RAIN
A sudden summer shower begins.
He looks toward the sky with mild surprise.
Instead of immediately running for shelter, he smiles and slowly walks into the rain.
The rain becomes heavier.
His hair becomes wet and falls naturally across his forehead.
He eventually starts running down the alley, laughing genuinely.
He briefly spins around while running, then continues toward a covered walkway.
His clothes become visibly damp.
Maintain realistic rain interaction, wet fabric, wet hair, reflections, and foot contact with the ground.
00:23–00:27 — QUIET MOMENT
He reaches the covered walkway and catches his breath.
Rain falls heavily behind him.
He wipes water from his forehead and looks quietly toward the street.
For a moment, everything becomes still.
He notices the camera again.
He gives a small genuine smile, not a posed expression.
00:27–00:30 — WALKING AWAY
The rain becomes lighter.
He walks away down the wet residential lane.
The camera follows from behind.
Reflections shimmer across the concrete.
He turns his head once, gives a tiny wave toward the camera, smiles, and continues walking.
The camera remains pointed toward the empty street for a brief moment.
At approximately 00:29, the recording abruptly cuts to black mid-motion.
No fade-out.
PHYSICAL REALISM
Maintain believable real-world physics throughout.
Hands, fingers, feet, clothing, hair, rain, bottle, basketball, and background objects must behave naturally.
No extra fingers, fused hands, duplicated limbs, distorted anatomy, floating objects, teleportation, disappearing objects, or sudden transformations.
The bottle remains a separate physical object and never intersects with his face.
The basketball behaves naturally and remains where it lands.
Parked bicycles and background objects remain stationary unless physically moved.
His feet remain properly connected to the ground while walking and running.
Keep the environment and subject consistent between shots.
AUDIO
Natural environmental audio only.
Footsteps on concrete, distant traffic, birds, leaves moving in the wind, bicycles, faint neighborhood conversations, shop sounds, bottle opening, basketball bouncing, rain hitting concrete, water dripping from rooftops, and subtle camera-handling noise.
No music.
No narration.
No soundtrack.
No artificial sound effects.
No spoken dialogue is necessary.
FINAL FEEL
The result should feel like a forgotten personal recording of an ordinary summer day.
Not a commercial.
Not a fashion film.
Not a music video.
Not a professional cinematic production.
The emotional appeal should come from small human moments: sitting alone, wandering through familiar streets, missing a basketball shot, drinking something cold, getting caught in the rain, laughing, and walking home.
Quiet, masculine, youthful, nostalgic, slightly melancholic, warm, spontaneous, and deeply human.
Prioritize natural behavior, consistent identity, believable physics, imperfect handheld framing, authentic environmental details, and the feeling that the camera happened to be there.
Use IMAGE 1 as the exact main character reference. Preserve the exact face, identity, hairstyle, clothing and body proportions.
Create a spectacular live-action martial-arts fight on the rooftop of a skyscraper. IMAGE 1 fights many attackers using extremely agile, playful and inventive martial-arts choreography: rapid footwork, evasions, pivots, spins, low sweeps, acrobatic rolls, vaults, controlled strikes and clever environmental interaction.
CAMERA CHOREOGRAPHY:
Start with a dramatic aerial establishing shot looking down between skyscrapers, then rapidly dive toward IMAGE 1.
Orbit 180 degrees around IMAGE 1 as several attackers surround him.
Use a low-angle tracking shot as IMAGE 1 ducks under an attack and spins past another enemy.
Whip-pan from one attacker to another as IMAGE 1 redirects their momentum and makes two attackers collide.
Use a fast side-tracking shot following IMAGE 1 as he runs across the rooftop.
Brief slow-motion only at the peak of a spectacular jump, then immediately return to full speed.
Camera performs a smooth 360-degree orbit during a spinning evasive sequence.
Finish with a long-lens push-in as IMAGE 1 suddenly sprints toward the rooftop edge.
FINAL SHOT:
Camera follows directly behind IMAGE 1 in a fast pursuit shot as he leaps from the skyscraper toward the suspended railway between two buildings.
Pull the camera backward into a huge wide aerial shot, revealing IMAGE 1 completely airborne above the enormous city gap.
End exactly while IMAGE 1 is reaching toward the suspended railway support.
Ultra-photorealistic live-action cinema, realistic physics, natural body weight, cinematic lighting, detailed city environment, controlled motion blur.
No supernatural abilities, no flying, no teleportation, no gore, no identity change, no costume change, no distorted anatomy, no excessive camera shake.
Create a 30-second, ultra-realistic cinematic product commercial for a fictional premium mirrorless digital camera called LENS ONE. Make it feel like a high-end global camera campaign with cinematic storytelling, beautiful photography, premium product cinematography and emotional everyday moments.
Feature a stylish young woman in her early 20s with long dark hair, natural makeup and a sophisticated outfit: cream fitted blazer, white top, high-waisted tailored trousers, minimal gold jewelry and elegant sneakers. Keep her face, hairstyle, body proportions and outfit consistent throughout.
0–4s: Open with a dramatic macro close-up of the LENS ONE resting on a premium dark surface. Slowly glide around the camera, revealing the textured grip, metallic body, lens barrel, control dial, shutter button and detailed lens glass. Show the aperture opening, then the woman naturally picking up the camera.
4–10s: She walks through a beautiful city street during golden-hour light and notices a child chasing soap bubbles. She raises the camera, looks through the viewfinder and presses the shutter. The exact moment briefly freezes into a beautiful cinematic photograph. Show the captured image clearly on the camera's rear display.
10–17s: She continues walking and photographs an elderly couple holding hands, a street musician and a dog running through sunlight. Include cinematic close-ups of her eye against the viewfinder, her finger pressing the shutter, the focus ring turning and the camera lens locking focus.
17–23s: Create a fast premium montage of the camera capturing different moments: a cyclist passing through golden light, rain droplets falling, friends laughing at a café and sunlight reflecting on wet pavement. Between shots, show detailed macro product shots of the lens, controls, shutter mechanism and rear display.
23–27s: She sits at an outdoor café and reviews the photographs on the camera screen. The images appear as a beautiful visual story of her day. She pauses on one photograph, smiles and looks back toward the street.
27–30s: She places the camera carefully on the table. The background becomes beautifully blurred as the camera becomes the sole focus. Slowly push toward the product, showing the lens glass catching warm sunset light and revealing the premium body details. End on a clean hero shot of the entire camera against the glowing city skyline.
Voiceover:
“Some moments happen once. Make them unforgettable.”
Ultra-realistic luxury camera advertising, cinematic bokeh, realistic metal and rubber textures, detailed lens reflections, natural skin and fabric textures, smooth camera movement, subtle film grain, realistic motion blur and authentic shutter sounds.
Keep the camera design, lens, buttons, dials, proportions and materials identical in every shot. No camera morphing, changing lens design, distorted hands, extra fingers, identity drift, duplicated people, unrealistic reflections, floating camera, excessive VFX, CGI appearance, subtitles, captions, random text, logos or watermark.
Create a 30-second ultra-realistic cinematic educational video set inside a modern university chemistry laboratory and lecture classroom. A confident male chemistry professor enters the room carrying a coffee cup and teaching materials while students prepare for class. Show the professor explaining chemistry concepts on a large digital board, with chemical equations, molecular structures, and a clear reaction diagram displayed behind him as students listen attentively.
Transition to a detailed close-up of a glass beaker containing two clear chemical solutions, highlighting the upcoming experiment and the principles of chemical reactions. Show the professor and students gathering around the laboratory table as he carefully mixes the solutions and demonstrates the reaction using realistic laboratory equipment. Capture natural hand movements, accurate pouring, mixing, bubbling, color changes, and realistic interactions with the glassware.
Show students reacting with curiosity and discussing the experiment while the professor explains the chemical changes they are observing. Include cinematic close-ups of the reacting solution, glass beaker, laboratory tools, the professor’s hands, and the students’ focused expressions.
Maintain scientifically realistic chemistry throughout, including accurate liquid behavior, mixing, chemical reactions, bubbles, temperature effects, and proper contact between materials, with no physically impossible results. Use smooth camera movements, shallow depth of field, realistic laboratory lighting, detailed facial expressions, and authentic university chemistry-lab surroundings.
End with the professor standing beside the completed experiment while the students observe the final result, creating a professional, inspiring, and photorealistic chemistry-education atmosphere.
A cinematic travel montage of a solitary young female hiker exploring vast dramatic natural landscapes. Photorealistic, ultra-high detail, filmic color science, anamorphic lens feel, subtle film grain, rich natural contrast. Exact same cinematography as the reference: contemplative wide frames, slow deliberate camera moves, epic scale that dwarfs the human figure, seamless ground-to-aerial transitions, dramatic weather and light changes.
Consistent subject
Young East-Asian woman, mid-20s, long straight dark hair, slender athletic build. Cream-white knit beanie → white bucket hat; light sage-grey technical rain jacket → white softshell hoodie; dark olive-green hiking backpack with red zipper accents; black hiking pants; trekking poles. Calm, solitary presence — never looks at camera, always absorbed in the landscape.
Shot-by-shot timeline (exact pacing & camera language)
0.0–1.8s
Wide static-to-slow push-in. Woman sitting on mossy green highland plateau, back to camera, hands resting on the grass, white beanie, sage jacket, dark backpack. Vast turquoise lake dotted with small rounded islands under heavy layered cloudy sky with soft blue openings. Soft overcast daylight, cool teal-green color grade. Gentle breeze moves her hair. Camera slowly creeps forward and slightly rises, emphasizing solitude and scale.
1.8–3.1s
Tracking shot from behind at medium-wide. She walks steadily uphill across rolling emerald hills, trekking pole in right hand, white beanie, same backpack. A vivid double rainbow arcs across the grey-green sky over distant soft mountains. Slight handheld micro-sway, camera following at her walking pace, low and slightly behind.
3.1–4.2s
Static wide shot. She stands alone on a small grassy spit at the edge of a dark alpine lake, facing away, white hoodie, yellow-tinged beanie. Towering jagged black peaks rise steeply, low white clouds and mist clinging to the rock faces. Moody, desaturated blue-grey light. Still water reflects the mountains. Pure contemplative silence.
4.2–5.9s
Slow aerial drone pull-back and slight orbit. High-angle reveal of the same dark lake from above — perfect circular dark water, dense conifer forest, tiny red-roofed building at the shore, dramatic mist-shrouded granite spires and snow patches. Low clouds pour over the ridges like a waterfall. Cool high-contrast desaturated grade. Smooth floating drone motion.
5.9–7.8s
Night sequence. Extreme wide. Black SUV with glowing red taillights drives slowly left-to-right on a flat dark lakeshore road. Massive violet-pink lightning bolt cracks across the deep indigo sky, illuminating distant mountains and water for a fraction of a second. Camera holds wide and static. Residual thin lightning forks linger after the flash.
7.8–9s
Golden-hour peak shot. Woman stands on a grassy ridgeline, white bucket hat, backpack, one trekking pole, looking left toward the horizon. Endless layered mountain ranges fade into soft atmospheric haze under glowing orange-pink sunrise/sunset. Soft rim light on her silhouette. Slow subtle push-in.
9–11.6s
Epic aerial tracking shot. Smooth high-speed drone follows a black SUV driving along a winding dirt mountain road through lush green karst-like hills. Multiple white wind turbines on ridges. Deep valleys with patchwork green fields, distant lake and town. Brilliant clear blue sky with scattered white cumulus. Camera flies parallel and slightly above the road, car becomes a tiny moving element.
11.60–14s
Intimate POV from inside a car (slightly soft foreground of A-pillar and side mirror). Through the open window we see the same woman standing outside on the overlook, white bucket hat, green backpack, holding binoculars to her eyes, gazing at the vast green mountain range and wind turbines under late-afternoon light. Soft focus on car interior, sharp on her and the landscape. Gentle breeze moves her hair.
Overall cinematography & technical notes
Camera language: locked-off contemplative wides, slow creeping push-ins, organic tracking, floating drone reveals, one intimate car-POV. Never shaky.
Lighting progression: soft overcast → dramatic rainbow → cold misty alpine → night lightning → golden hour → crisp clear daylight.
Color science: cool teal-greens and greys early, warm golden tones mid-way, rich emerald and deep blue later. Subtle anamorphic flare and gentle film grain.
Mood: quiet awe, solitude, the scale of nature vs one small human traveler. 24 fps cinematic motion, 4K detail.
Create a 15-second vertical 9:16 ultra-realistic cinematic action sequence set on a massive outdoor stage in a futuristic neon city, inspired by a high-budget fighting game.
A huge muscular male fighter with spiky golden hair, wearing loose white martial-arts pants, a red waist sash, golden wristbands and boots, stands confidently in the center of the stage while a massive crowd watches and records on their smartphones. Behind him is a dense modern city skyline filled with glowing billboards, neon signs, skyscrapers, smoke, and dramatic evening lighting.
A blonde female fighter wearing a dark blue and red superhero-style combat suit suddenly attacks him at high speed. She leaps through the air and delivers a powerful spinning kick toward the male fighter. He remains composed, then counters with an explosive martial-arts strike, sending her flying backward across the stage.
Show dynamic hand-to-hand combat with realistic body mechanics, powerful impacts, flying dust and smoke, subtle camera shake, dramatic slow motion during the kick, and a fast cinematic camera push-in during the counterattack.
The crowd reacts with excitement, raising their phones and cheering. Capture the atmosphere of a live street-fighting spectacle with hundreds of spectators in the foreground.
Visual style: photorealistic AAA video-game cinematic, highly detailed characters, realistic skin and fabric textures, physically accurate lighting, volumetric smoke, neon reflections, cinematic depth of field, realistic motion blur, dramatic contrast, dynamic camera movement, epic action choreography.
Camera: vertical handheld smartphone-style perspective mixed with cinematic tracking shots, low-angle hero shots, rapid push-ins, slight camera shake during impacts, smooth slow motion at key moments.
End with the female fighter being thrown into the air as smoke and dust fill the stage, while the muscular fighter stands triumphantly in the center and the crowd erupts behind him.
No dialogue, no subtitles, no text overlays, no logos, no watermark. Maintain consistent character appearance throughout the entire sequence.
Create a 15-second live-action cinematic samurai battle, 16:9, ultra-realistic movie quality, practical-looking effects, natural human movement, detailed costumes, realistic physics, and dramatic handheld camera work. Avoid an obvious AI or cartoon appearance.
Story: After protecting a sacred village, a wounded white-haired samurai returns to an ancient torii gate at sunset—only to find the demon warrior who destroyed his clan waiting for him. This is their final confrontation.
0–3s: Wide establishing shot. A ruined Japanese shrine and massive torii gate stand beneath a blood-red sunset. Dust blows across the ground. The samurai slowly walks forward, gripping his black katana. Across from him, a huge dark armored demon waits silently, red eyes glowing through the smoke.
3–6s: The demon suddenly charges. The samurai blocks a devastating strike. Their swords collide repeatedly—fast, heavy, and physical. Sparks burst from every impact as they move through the ruined shrine.
6–11s: The fight becomes relentless. Close-range sword exchanges, brutal parries, spinning attacks, dodges, low slides, and powerful counterattacks. The demon smashes through wooden structures while the samurai narrowly escapes. Camera rapidly follows them with realistic motion blur, impact shakes, flying dust, sparks, and debris. The samurai finally finds an opening.
11–13s: In one explosive final exchange, both warriors sprint toward each other. Their blades cross in slow motion. The samurai lands a precise finishing cut and stops behind the demon.
13–15s: Silence. The demon collapses into drifting black ash. The samurai lowers his sword and looks toward the enormous setting sun. Wind moves his torn haori as the last embers disappear into the sky.
Visual style: Photorealistic live-action Japanese period film, grounded martial-arts choreography, realistic sword physics, practical-looking smoke and sparks, natural skin and fabric texture, cinematic lighting, deep shadows, anamorphic lens feel, subtle film grain, realistic camera movement, dramatic pacing, believable weight and impact.
Character consistency: One white-haired wounded samurai in a torn white haori and one towering black-armored demon with glowing red eyes. No extra characters, no duplicates, no subtitles, no text, no watermark, no cartoon look, no exaggerated CGI, no artificial-looking movement.
A cinematic, surreal AI-generated video featuring a regal woman in an elegant gown mounted on a horse atop the precipice of a towering mountain cliff, overlooking expansive valleys beneath dramatic sky formations. Attached to her dress is an extraordinarily long, translucent silk train that billows dynamically in high-altitude wind, sweeping up toward the sun and blending into voluminous, fluffy clouds. The visual sequence transitions through majestic landscapes—from jagged ridges at golden hour to misty volcanic peaks—while the sweeping fabric shifts in vivid hues of crimson, cobalt blue, warm gold, and pure white, creating an ethereal, dreamlike atmosphere with cinematic lighting and hyper-realistic detail.
Create a 15-second ultra-realistic cinematic UGC-style video where a young female creator suddenly finds herself physically inside a fictional open-world video game. She is a real, live-action human character interacting naturally with the game's world, not a gamer watching a screen.
CHARACTER CONSISTENCY:
The exact same young female creator throughout the entire video. Clear realistic face, natural skin texture, expressive eyes, stylish Gen-Z streetwear, same hairstyle and outfit in every shot. She must look like a real person seamlessly placed inside the game world.
0–3 SEC — ENTERING THE GAME
Start with the creator standing in a realistic bedroom holding a game controller. The screen suddenly glitches with a dramatic digital distortion. In a seamless transition, the environment around her transforms and she is teleported directly into a huge fictional open-world city.
She looks around in shock and says:
“Wait… I’m actually inside the game?!”
3–7 SEC — OPEN-WORLD CHAOS
She is now physically walking through the game's city streets as a live character. Massive skyscrapers, pedestrians, sports cars, traffic lights, shops and cinematic city details surround her.
She looks at her hands, then touches a nearby car and reacts with genuine disbelief.
A police siren suddenly starts in the distance.
7–11 SEC — THE CHASE
She jumps into a fictional sports car and starts driving through the city.
Camera switches between third-person cinematic game-camera shots and close-up live-action shots of her face inside the vehicle.
Police cars begin chasing her. She laughs nervously and says:
“Oh no… I think I’m wanted!”
11–15 SEC — ESCAPE
She dramatically turns into a narrow alley, narrowly avoids a pursuing police vehicle, and escapes into an underground parking garage.
She stops the car, looks directly into the camera and says:
“Okay… how do I get OUT?!”
End with a dramatic cinematic pull-back revealing the enormous fictional game city around her.
VISUAL STYLE
Ultra-realistic live-action human seamlessly integrated into a photorealistic open-world game environment, cinematic game graphics, realistic physics, natural facial expressions, detailed city, dynamic camera movement, realistic vehicles, dramatic lighting, shallow depth of field, high-end VFX, premium gaming trailer + authentic UGC reaction style.
IMPORTANT: The woman must remain a realistic live human character inside the game world. Do not make her look like a cartoon, CGI avatar, animated character, or video-game model.
Create an original fictional open-world crime-action game aesthetic. Do not use GTA logos, characters, locations, UI, music, copyrighted assets, or recognizable game elements.
No text overlays, subtitles, watermarks, logos, or futuristic objects.
Create a 30-second, 1080p ultra-realistic cinematic product commercial for fictional premium smart water bottle LUMEN. The bottle should be the hero product while the woman remains visible and naturally interacts with it throughout the film.
Character consistency: Same stylish young woman, early 20s, long dark hair, natural makeup and identical facial features throughout. Keep the exact LUMEN bottle design, color and proportions consistent.
Scene 1 — City Morning
The woman walks through the city wearing a fitted white athletic top, high-waisted beige trousers, white sneakers and a lightweight cream jacket, carrying the LUMEN bottle. Show her full body, then a brief close-up of the bottle and hydration indicator.
Scene 2 — Park Workout
She arrives at a park wearing a black fitted sports top, black high-waisted leggings and white running shoes. After jogging, she takes a refreshing sip from LUMEN. Show condensation, water movement and the bottle’s premium materials.
Scene 3 — Café
She sits at a modern café wearing a soft beige knit sweater, straight-leg blue jeans and white sneakers, with LUMEN beside her. She picks it up and drinks while working on her laptop.
Scene 4 — Visual Transformation
As she takes another sip, transition into cinematic visuals of crystal-clear mountain water, flowing streams and ocean waves, then seamlessly return to the bottle in her hand. Keep her outfit consistent with the café scene.
Scene 5 — Sunset Hero Shot
She stands on a rooftop wearing a sleek black blazer, fitted white top, tailored trousers and elegant sneakers, holding LUMEN while watching the sunset. Camera slowly pushes toward the bottle, capturing reflections, condensation and the illuminated hydration indicator.
Voiceover:
“Your day moves fast. Your hydration should keep up. LUMEN keeps every sip within reach and reminds you when it matters. Stay refreshed. Stay moving.”
Style: Photorealistic luxury product cinematography, realistic skin and fabrics, cinematic lighting, macro product details, natural movement, shallow depth of field, realistic reflections, smooth transitions and premium commercial quality.
Negative: No bottle morphing, design changes, identity drift, distorted hands, extra fingers, random outfits, duplicated people, floating objects, excessive CGI, subtitles, captions, random text or watermark.
Create a 20-second ultra-realistic cinematic UGC advertisement, vertical 9:16, featuring the EXACT SAME male character from Reference Image 1. Strictly preserve his facial identity, face structure, skin tone, hairstyle, hairline, eyes, nose, lips, jawline, beard/stubble and overall appearance. No identity drift, face morphing, aging or redesign.
PRODUCT: premium modern wireless earbuds with sleek minimalist matte-black charging case, subtle metallic details, realistic luxury-tech appearance, no recognizable brand/logo.
WARDROBE: fitted black premium T-shirt, minimalist dark overshirt/jacket, black watch, subtle accessories, sophisticated urban look.
SETTING: modern luxury apartment/stylish urban workspace during golden hour, warm window light, realistic shadows, tasteful furniture and subtle city ambience. Premium lifestyle + authentic UGC + cinematic commercial aesthetic.
0–4s: Close-up of male looking slightly tired after a busy day. He picks up the earbud case from a table, camera quickly pushes toward product. He looks at camera confidently: “These changed my daily routine.”
4–8s: Cinematic macro reveal as he opens the case, removes an earbud and puts it in. Smooth product movement and shallow depth of field. Dialogue: “The sound is seriously impressive.”
8–13s: Male walks through apartment and works at desk wearing earbuds. Show natural tap interaction, earbud close-up, working shots and product case. Dialogue: “Music, calls, work… everything feels better.”
13–17s: Male stands by window in golden-hour light, looks directly into camera and holds case toward lens. Dialogue: “Honestly, I keep them with me everywhere.”
17–20s: Hero product shot: earbuds and case sharply focused in foreground, male softly blurred behind. Slow cinematic movement and premium reflections. On-screen text: “Sound. Style. Freedom.” then “Upgrade Your Everyday.”
AUDIO: original modern premium lifestyle beat with smooth electronic bass, atmospheric synth and subtle percussion; minimal at start, gradually builds, slight impact on final reveal. No copyrighted music. Dialogue remains crisp.
Natural American English male UGC voice, confident, relaxed, authentic, conversational, accurate lip-sync.
Photorealistic 4K/8K, realistic skin, eyes, blinking, hands, fabric, reflections, shadows, depth of field and motion blur. No distorted hands, extra fingers, text glitches, random logos, watermarks or cartoon look. Maintain exact identity and clothing consistency throughout.
Create a ultra-realistic cinematic brand commercial for a fictional premium electric scooter brand called VOLTIS. The film follows a young woman moving through a modern city, showing how VOLTIS turns an ordinary commute into an effortless urban adventure.
Character: Same young woman throughout, long dark hair, natural makeup, confident and energetic personality.
Scooter: Premium matte-black electric scooter with subtle metallic details, integrated LED headlight, digital display, wide deck, sleek aerodynamic frame and clearly visible VOLTIS branding. Keep the exact scooter design, logo, proportions and details consistent throughout.
Scene 1 — The Start | 0–5s
Early morning. The woman leaves a modern apartment wearing a cream fitted jacket, white top, straight-leg black trousers and white sneakers. She unlocks the VOLTIS scooter and checks the digital display.
Close-up of the VOLTIS logo, handlebar controls and LED display.
VO:
“Somewhere between here and there… there’s a better way to move.”
Scene 2 — City Commute | 5–10s
She rides through a clean modern city street wearing the same outfit. Use smooth tracking shots, front and side angles and a few wide shots showing the entire city around her.
She effortlessly moves through the morning commute while cars remain stuck in traffic.
Show the scooter's smooth acceleration and responsive handling.
Scene 3 — Different Terrain | 10–15s
Transition to a city park and smooth pedestrian cycling path. She wears a lightweight sage-green windbreaker, black athletic leggings and the same white sneakers.
She rides through the park, passes trees and open spaces, then takes a gentle uphill path.
Show close-ups of the wheels, suspension and deck moving naturally over the pavement.
VO:
“Smooth when the city moves fast. Steady when the road changes.”
Scene 4 — Unexpected Rain | 15–20s
Clouds gather and rain begins.
She now wears a black waterproof jacket, black leggings and the same sneakers. She continues riding carefully along wet urban streets.
Show realistic rain droplets, wet pavement reflections, water splashes from the tires and the LED headlight illuminating the road.
VO:
“Because your journey doesn’t always go according to plan.”
Scene 5 — Sunset Freedom | 20–26s
Golden hour. She rides along a scenic riverside road wearing a beige oversized jacket, white top, relaxed dark trousers and white sneakers.
The city glows behind her. Use a wide cinematic tracking shot followed by a low-angle shot of the scooter moving smoothly along the road.
She smiles naturally as she rides.
Scene 6 — Final Hero Shot | 26–30s
She stops at a beautiful city overlook at sunset and steps off the scooter.
She looks toward the skyline while standing beside VOLTIS.
Slow camera push-in toward the scooter. The VOLTIS logo and sleek silhouette are clearly visible.
VO:
“VOLTIS. Take the shortcut.”
End on a clean premium brand frame featuring the scooter and VOLTIS logo.
Visual Style
Premium global mobility campaign, photorealistic live-action aesthetic, cinematic city cinematography, realistic riding physics, natural human movement, dynamic tracking shots, realistic reflections, subtle motion blur, golden-hour lighting, detailed materials, authentic urban environments and sophisticated commercial pacing.
Safety: Show the rider wearing a properly fitted helmet whenever riding, with controlled riding in appropriate areas and no dangerous traffic maneuvers.
Negative prompt: scooter design changes, misspelled VOLTIS logo, logo distortion, unrealistic riding physics, floating wheels, distorted hands, extra limbs, dangerous traffic behavior, excessive CGI, identity drift, inconsistent outfits within scenes, duplicated people, random text, subtitles, watermark.
Create a premium 15-second cinematic commercial for a global lifestyle brand. Start with a dramatic close-up of the product in a dark, luxurious environment, with soft cinematic lighting and subtle atmospheric particles. At 3 seconds, smoothly transition into dynamic macro shots showing the product’s premium design and details. From 7–11 seconds, show a stylish young professional confidently using the product in a modern international city environment, with elegant camera movement and shallow depth of field. At 11–15 seconds, return to a hero product shot with a clean background, dramatic rim lighting, and space for the brand logo and tagline. Ultra-realistic, high-end commercial cinematography, premium color grading, smooth transitions, realistic reflections, 4K detail, sophisticated international advertising aesthetic, no extra text, no watermark.
Create a 30-second ultra-realistic professional vlog video, filmed like authentic footage from a real travel/content creator visiting a modern wildlife zoo. 16:9 horizontal, 4K, natural daylight, realistic skin texture, real-world physics, authentic animal behavior, natural camera movement, realistic environmental audio, professional documentary/vlogging quality.
The entire video must feel 100% real and naturally filmed, not CGI, not animation, not a cinematic fantasy scene, and not AI-generated. Avoid overly perfect lighting, plastic-looking skin, unnatural animal movements, fake-looking backgrounds, excessive sharpness, or impossible camera movements.
MAIN VLOGGER:
A young adult woman in casual modern clothes, natural makeup, realistic hair movement, carrying a small vlog camera. Keep her face, hairstyle, clothing and appearance perfectly consistent throughout all shots.
0–5 SEC — HOOK
Start with a handheld selfie shot as the girl walks through the zoo entrance.
She looks directly into the camera with genuine excitement.
Girl:
“Okay, I just got inside the zoo… and apparently, the lions are this way!”
Natural footsteps, distant visitors, birds and subtle zoo ambience in the background.
Quick natural camera movement as she turns the camera toward the animal section.
5–10 SEC — WALKING THROUGH THE ZOO
Cut naturally to a walking shot following her from slightly behind.
She walks along a realistic zoo pathway surrounded by trees, signs, families and visitors.
She occasionally looks toward animal enclosures.
Girl:
“I honestly didn’t expect it to be this busy today.”
Her voice should sound spontaneous and conversational, with natural pauses and breathing.
10–16 SEC — LION ENCLOSURE
She reaches a realistic lion enclosure.
Switch to a handheld over-the-shoulder shot showing her watching a real-looking adult lion resting naturally on a large rock.
The lion moves naturally, blinks, breathes, looks around and briefly raises its head.
No exaggerated animal behavior.
Girl, quietly:
“Oh wow… look at him. He’s literally just chilling.”
Natural zoo ambience continues.
16–22 SEC — ELEPHANT AREA
Smooth but realistic handheld transition to the elephant enclosure.
A large elephant naturally walks toward a water area, uses its trunk to drink and gently moves its ears.
The girl appears beside the enclosure, keeping a safe distance.
Girl:
“And the elephants are even bigger than I imagined.”
She laughs naturally.
22–27 SEC — CLOSE VLOG MOMENT
Return to a selfie shot.
The girl walks slowly while talking directly to the camera.
Sunlight naturally filters through the trees.
Girl:
“I think this might already be my favorite part of the day.”
She briefly turns the camera toward the animals behind her.
27–30 SEC — NATURAL ENDING
End with a spontaneous shot of the girl smiling while walking away from the camera.
She turns back toward the lens.
Girl:
“Anyway, I’m going to explore the rest of the zoo. Come with me!”
She laughs slightly and continues walking.
REALISM & PRODUCTION REQUIREMENTS
- Photorealistic live-action appearance
- Authentic zoo environment
- Realistic human skin, hair and facial expressions
- Natural animal anatomy and behavior
- Realistic animal fur and skin textures
- Natural daylight and shadows
- Realistic reflections and materials
- Handheld vlog camera movement with subtle stabilization
- Occasional natural autofocus adjustment
- Realistic exposure changes when moving between bright and shaded areas
- Authentic background visitors behaving naturally
- Natural ambient zoo sounds
- Girl's voice recorded like a real vlog microphone
- Accurate lip synchronization
- Natural blinking, breathing and body movement
- No exaggerated cinematic effects
- No fantasy elements
- No artificial-looking animals
- No distorted faces or hands
- No duplicated people
Create a realistic 20-second cinematic travel vlog featuring one young female travel vlogger exploring a beautiful city. 0–5 sec: She walks through a lively street, holding a phone/camera and says, “Hey guys, today I’m exploring this beautiful place!” 5–10 sec: She stops at a small local food stall, smiles at the camera, tries a local snack and says, “Wow, this is really good!” 10–15 sec: She continues walking through a colorful market, showing the surroundings and briefly talking to the camera. 15–20 sec: She reaches a scenic viewpoint, looks around happily and says, “What an amazing day!” then walks forward toward the next destination. Natural handheld camera movement, authentic vlog style, realistic facial expressions, natural lighting, smooth transitions, ambient city sounds, consistent character and outfit throughout, highly realistic 4K video.
Create a 15-second ultra-realistic cinematic fight scene, shot like a high-budget action movie. The scene takes place at night in a narrow urban street after light rain. The wet asphalt reflects streetlights and neon signs, creating realistic cinematic reflections. Two adult male fighters face each other, both wearing slightly dirty, realistic everyday clothing.
0–3 seconds: Start with a wide establishing shot. Both fighters stand several feet apart, breathing heavily and staring at each other with intense expressions. Light rain falls around them. The camera slowly pushes forward with subtle handheld movement. Realistic ambient city sounds, distant traffic, footsteps, rain and heavy breathing.
3–6 seconds: One fighter suddenly moves forward and throws a realistic punch. The other fighter quickly blocks and steps backward. Use natural body mechanics, realistic weight, believable footwork and authentic reactions. Camera switches to a dynamic medium tracking shot, following the movement.
6–10 seconds: The second fighter counters with two fast punches. The first fighter dodges the first punch and blocks the second. Show realistic impact reactions without exaggerated superhero movements. Clothing and hair react naturally to the motion. Use a brief slow-motion moment during the dodge, then immediately return to real-time speed.
10–13 seconds: The fighters exchange a short sequence of close-range defensive movements and strikes. Keep the choreography grounded and realistic, with natural balance, fatigue and imperfect movements. Camera moves around them in a smooth handheld cinematic arc.
13–15 seconds: One fighter lands a strong final body-level strike, causing the other fighter to stumble backward and fall against a wall. End with a close-up of the standing fighter breathing heavily while rain continues falling. The camera slowly pulls back as the street lights flicker in the background.
Visual style: photorealistic, cinematic movie quality, realistic human anatomy, natural skin texture, realistic facial expressions, physically accurate movement, detailed wet clothing, realistic rain droplets, atmospheric night lighting, subtle lens flare, shallow depth of field, natural motion blur, realistic shadows, high dynamic range, cinematic contrast, 35mm film look, professional action cinematography.
Camera: realistic handheld camera, smooth tracking shots, wide establishing shot, medium action shots, close-up reaction shots, natural camera shake during impacts, realistic focus pulls, cinematic depth of field.
Lighting: practical streetlights, soft neon reflections on wet pavement, realistic rim lighting, subtle volumetric atmosphere, natural shadows, no artificial glowing effects.
Audio: realistic rain, footsteps on wet asphalt, breathing, clothing movement, distant traffic, realistic impact sounds, subtle cinematic background score that builds tension.
Important: Make the entire sequence feel like real footage captured by a professional film crew, not animation or a video game. No superhero physics, no exaggerated punches, no impossible movements, no cartoon effects, no excessive camera shake, no distorted faces, no extra fingers or limbs, consistent characters throughout the entire 15 seconds.
WAN 3.0 IMAGE-TO-VIDEO | ALEXA GREY | 30-SECOND BTS COMMERCIAL
REFERENCE:
Use the attached Alexa Grey Character Sheet as PRIMARY identity reference and storyboard as SECONDARY visual reference. Preserve Alexa’s exact face, facial proportions, warm deep-brown skin, dark brown eyes, long wavy black hair, gold hoop earrings, layered gold necklaces and athletic/hourglass proportions. SAME WOMAN, SAME FACE, SAME IDENTITY throughout. Keep her orange beach outfit consistent.
PRODUCT LOCK:
One single orange juice energy drink can. Keep identical shape, size, orange/yellow packaging, label, artwork and material throughout. Same hand holding the product. No switching hands, redesign or deformation.
STYLE:
Photorealistic premium beverage commercial filmed as authentic BTS footage. Tropical beach, wet sand, ocean waves, palm trees, warm golden-hour sunlight, cinematic depth of field, realistic skin/hair and natural handheld camera movement. Alexa is confident, playful and spontaneous, not overly rehearsed.
IMPORTANT:
Alexa SPEAKS every line marked ALEXA. Generate accurate natural English lip-sync. Director is OFF-SCREEN only. No subtitles.
00:00–00:04
CAM: Handheld side tracking, starting low and tilting from her legs/waist toward face.
ACTION: Alexa walks along shoreline holding the drink naturally. She looks around at the ocean.
ALEXA: “Wow… this place is beautiful.”
00:04–00:08
CAM: Smooth front-side arc into medium close-up.
ACTION: Alexa notices the product is too low, raises it herself and turns the label slightly toward camera.
ALEXA: “Okay, let’s make this look good.”
DIRECTOR OFF-SCREEN: “Perfect. Just relax and be yourself.”
00:08–00:12
CAM: Medium close-up, subtle handheld push-in.
ACTION: Alexa gives an exaggerated commercial pose, immediately realizes it feels forced and laughs, covering her mouth.
ALEXA: “Okay… that was definitely too much.”
DIRECTOR: “Exactly! Keep that laugh.”
00:12–00:16
CAM: Lower tracking movement, rising naturally with her.
ACTION: Alexa lowers her hand, brushes her hair back and relaxes. A small wave reaches her foot. She looks down, surprised, then laughs.
ALEXA: “Oh! The ocean wants a cameo too.”
DIRECTOR: “Forget the pose. Just be Alexa.”
00:16–00:20
CAM: Follow laterally into a relaxed three-quarter angle.
ACTION: Alexa looks toward the ocean, takes a natural breath, then glances back at camera. Her expression becomes calm and confident.
ALEXA: “Honestly… this is more my kind of energy.”
00:20–00:24
CAM: Medium close-up framing Alexa and product, subtle push-in.
ACTION: Alexa rotates the can naturally so the front faces camera. She studies it for a moment, then smiles.
ALEXA: “Natural energy. Real juice.”
DIRECTOR: “Yes. That’s the shot.”
00:24–00:28
CAM: Smooth waist-up hero tracking shot.
ACTION: Alexa walks slowly toward camera, drink clearly visible. Hair moves naturally in the sea breeze. She looks directly into lens.
ALEXA: “Fresh energy… without trying too hard.”
00:28–00:30
CAM: Slow cinematic push-in to final hero close-up.
ACTION: Alexa stops, raises the drink slightly and gives a confident, genuine smile.
ALEXA: “Fresh energy. No pretending.”
DIRECTOR OFF-SCREEN: “Beautiful. That’s a wrap.”
END with Alexa and the orange juice energy drink clearly visible against the ocean sunset.
AUDIO:
Natural ocean surf, sea breeze, footsteps, clothing movement and subtle beach ambience. Premium upbeat commercial music gradually builds. Alexa’s voice is warm, conversational and confident. Director remains off-screen.
CAMERA:
Use handheld tracking, low angles, medium close-ups, subtle arcs, push-ins, three-quarter angles and smooth hero movement. Realistic motion blur, shallow depth of field, cinematic lensing and natural golden-hour lighting. No static shots.
PRODUCT:
The same orange juice energy drink must remain consistent in every frame. Keep packaging readable and unchanged. Final frames prioritize product visibility.
NEGATIVE:
NO face change, NO identity drift, NO duplicate Alexa, NO extra people, NO hairstyle change, NO skin-tone change, NO body distortion, NO extra fingers, NO malformed hands, NO hand switching, NO product deformation, NO packaging change, NO product duplication, NO outfit change, NO visible director, NO subtitles, NO gibberish, NO misspelled text, NO exaggerated acting, NO drinking, NO pouring, NO jump cuts, NO camera glitches, NO warped background.
FINAL FEEL:
A believable premium orange juice energy drink commercial that feels like genuine BTS footage from a real shoot. Alexa’s personality drives the commercial: confident, funny, natural and relatable. SAME ALEXA from first frame to last.
SCENE CONTEXT
Premium 30-second Japanese cherry juice commercial: playful, surreal, fashion-forward, photorealistic and highly polished. Energetic pacing, Japanese fashion-ad aesthetics, creative HARD CUTS, MATCH CUTS and WHIP CUTS.
ACTIVE REFERENCE
<<<image_1>>> is the exact SIPPO Cherry carton reference. Preserve its matte powder-pink rectangular carton, folded top, white flexible straw, striped sunset symbol, dark-red SIPPO logo, Cherry label and two smiling cherries. Never change the packaging into another container.
CHARACTERS
Four adult Japanese women, 22–28, distinct hairstyles: black bob, raven ponytail with burgundy ribbons, copper-brown bob, waist-length black hair with blunt bangs. Tokyo fashion styling in cherry red, pink, ivory and chrome. Stable faces, natural skin and believable friendship.
FIRST FRAME
Carton already dominates the foreground with its full front panel readable; lead woman directly behind it. No establishing shot.
FORMAT MODE
Fast multi-shot commercial with consistent product, characters, wardrobe and color world. Every shot introduces a new visual idea.
ACTION AND CAMERA SEQUENCE
OPENING — 107° wide rectilinear shot. Carton fills foreground on glossy pink table. Lead woman smiles and points at straw as camera rapidly pushes in.
WHIP CUT — Camera bursts through a cherry-red circular portal into a surreal Tokyo studio street with pink tiles, cream walls, chrome rails and giant cherries. Four women walk toward camera in synchronized formation. 84° wide stabilized dolly.
MATCH CUT — Carton passes through a circular wall opening and emerges into another woman’s hands inside a burgundy/pink Japanese photo booth. Fast 180° orbit as she sips and side-eyes camera.
OVERHEAD — Four women form a radial composition around a cherry-red platform, passing the carton clockwise while camera rotates opposite. Two cherries roll across frame as transition.
MACRO INSERTS — Condensation, flexible straw, fingertips, glossy cherry, SIPPO logo and smiling cherry graphics. Product remains faithful to <<<image_1>>>.
LOW ANGLE — Four women surround the product pedestal. Camera rapidly cranes upward as they turn toward camera and break into natural laughter. Realistic hair and fabric motion.
FINAL GROUP — 84° wide cherry-red stage with giant circular sunset-inspired light. Each woman holds an identical carton and they perform a synchronized toast.
FINAL PACKSHOT — HARD CUT to cream-pink studio. Exact SIPPO Cherry carton on polished pedestal with two cherries beside it. Straw bends right. Slow premium telephoto push. Hold for brand recognition.
LIGHTING AND IMAGE QUALITY
Premium Japanese beauty-commercial lighting, soft overhead light, cherry-red edge light, pearlescent skin, powder pink/burgundy/cream palette. Photorealistic ARRI Alexa 35 look, natural grain, HDR, pristine commercial finish.
PHYSICS
Realistic carton weight and stiffness, natural finger pressure, flexible straw, gravity-driven condensation, believable hair, fabric and cherry motion.
AUDIO
Original upbeat Japanese electro-pop with female vocal chops, bass, claps and sparkling synths. Add synchronized carton taps, straw flex, cherry rolls, fabric swishes, whooshes and sip sounds. Final toast: “SIPPO Cherry!” End with a two-note sonic logo. No subtitles.
Create a 10-second vertical 9:16 ultra-photorealistic luxury beauty commercial following all 18 storyboard shots in order. Keep one identical blonde model with messy tied hair, blue-gray eyes, freckles, glossy lips, natural skin and an off-shoulder Tiffany Blue dress. All nails are glossy turquoise, medium almond-shaped. One identical transparent glass nail-polish bottle with turquoise lacquer and silver cap. Pure white studio, soft beauty lighting, realistic textures, cinematic macro, smooth orbits, focus pulls and elegant match cuts. 0–0.5s: Extreme macro of glossy turquoise nails, slow orbit. 0.5–1s: Pull back revealing the same hand holding turquoise ice cream in a waffle cone. 1–1.5s: Reveal model behind ice cream, playful eye contact, focus shifts to face. 1.5–2s: Macro of two manicured fingers holding polish bottle, subtle push-in. 2–2.5s: Brush emerges from bottle with a realistic turquoise lacquer droplet. 2.5–3s: Extreme macro of brush smoothly applying glossy polish to one nail. 3–3.5s: Model playfully peeks from behind a giant turquoise macaron, focus pull to eye. 3.5–4s: Manicured fingers place a macaron onto a stack of three, tactile macro. 4–4.5s: Beauty portrait, model touches cheek with turquoise nails, slow push-in. 4.5–5s: Extreme nail macro transitions to matching turquoise macaron texture. 5–5.5s: Model playfully brings ice cream near lips, rotates cone, camera arcs 45 degrees. 5.5–6s: Both manicured hands elegantly frame polish bottle, slow macro push-in. 6–6.5s: Fingers lift a glossy turquoise ceramic cup, focus on nails. 6.5–7s: Close-up of glossy lips approaching turquoise macaron, rack focus to nails. 7–7.6s: Top-down shot of hand spreading five perfect turquoise nails across white surface. 7.6–8.2s: Model holds polish bottle at eye level, focus shifts from product to eyes. 8.2–9s: Overhead composition of polish bottle, turquoise macarons, ice cream and cup; manicured hand picks up bottle, gentle orbit. 9–10s: Final beauty hero portrait, model holds bottle beside face while other hand frames cheek, all nails visible, slow push-in and final hold. Maintain exact identity, colors, realistic hand anatomy, natural movement and product consistency. No extra people, objects, text or overlays
Extremely fast paced, realistic, cinematic FPV flying through Disneyland. Low altitude over Sleeping Beauty Castle, parade streets, and fantasy villages. Sharp dives past rollercoasters, spinning teacups, and fireworks launch zones. Gliding above rivers with boats, glowing lights, and crowded themed lands. Realistic textures, reflections, dynamic shadows, steam, and smooth fluid movement. Close passes through tunnels, animatronic sets, and neon-lit rides.
Create a 30-second ultra-realistic open-world GANGSTER gameplay cinematic featuring two rival street gangs fighting in a gritty modern city at night. Make the entire video feel like a trending AAA gameplay recording, as if a player is actively controlling the main character.
GAMEPLAY SCREEN / HUD — VERY IMPORTANT:
Add a polished, original modern gameplay interface that stays visible naturally throughout the video:
Top-left: small circular character portrait + green HEALTH BAR and blue STAMINA BAR
Top-right: compact MISSION OBJECTIVE panel
Bottom-left: detailed MINIMAP, circular radar, player arrow, enemy indicators, nearby vehicle icons and objective marker
Bottom-right: transparent action-button icons for punch, dodge, sprint, interact and crouch
Center: small dynamic crosshair/aim indicator when appropriate
Enemy targets: red diamond markers above hostile characters
Teammates: blue/green indicators above friendly characters
Combat feedback: subtle hit indicator, combo counter and XP/reward notification
Mission notifications: sleek pop-up messages such as “RIVAL GANG DETECTED”, “SURVIVE THE AMBUSH”, and “BOSS FIGHT INCOMING”
Add subtle controller-style button prompts when the player approaches objects or enemies
HUD should be semi-transparent, futuristic and minimal, integrated naturally into the scene rather than covering the action.
UI animations should feel responsive: icons pulse, objectives update, health decreases realistically, enemy markers move with characters.
Make the interface original and not an exact recreation of any existing game UI.
0–5 SEC — RIVAL GANG DETECTED
Third-person camera follows the protagonist walking through a neon-lit city street.
HUD:
Minimap activates → several red enemy markers appear → objective panel displays:
MISSION
RIVAL GANG DETECTED
An enemy warning icon flashes briefly.
5–10 SEC — STREET FIGHT
A rival gangster attacks. The player-controlled protagonist blocks, dodges and counterattacks.
Show bottom-right action icons responding to the player's movements. Red enemy markers appear above opponents. Add a small COMBO ×3 indicator after consecutive successful moves.
10–15 SEC — CHAOS
The fight spreads across the street. Gang members run between cars and barriers.
The minimap becomes crowded with moving red and blue indicators.
OBJECTIVE UPDATED:
SURVIVE THE AMBUSH
Add subtle screen shake and damage feedback when the protagonist gets hit.
15–20 SEC — SLOW-MOTION COUNTER
The protagonist becomes surrounded. Gameplay suddenly enters dramatic slow motion.
A DODGE prompt appears near the bottom-right controls. He dodges an incoming attack and performs a counterattack.
The HUD smoothly returns to normal speed.
20–26 SEC — GANG WAR
Both gangs charge toward each other.
The protagonist runs through the battlefield while teammates fight alongside him.
OBJECTIVE:
DEFEAT THE RIVAL GANG
Show dynamic enemy markers, teammate indicators, minimap movement, stamina depletion while sprinting, and realistic combat feedback.
26–30 SEC — BOSS FIGHT
A black SUV arrives. The rival gang leader steps out.
All enemy indicators converge around him.
A dramatic BOSS HEALTH BAR appears across the top-center:
RIVAL GANG LEADER
████████████████████
Then a large original gameplay notification appears:
⚠ BOSS FIGHT INCOMING
The protagonist raises his fists. Camera moves behind him into a classic third-person gameplay position.
CUT TO BLACK.
Create a 30-second photorealistic smartphone travel vlog featuring a young Turkish woman exploring Istanbul on a warm summer evening. She wears a cute, elegant cream-colored linen dress with delicate embroidered details, a light patterned scarf loosely tied around her hair, small gold earrings, and a woven shoulder bag. Her look feels stylish, natural, and effortlessly Turkish.
The vlog opens in selfie mode as she walks through a lively historic Istanbul street, smiling at the camera while colorful shop signs, old Ottoman-style buildings, hanging lamps, and passing locals create an authentic atmosphere. She turns the camera toward a small street vendor and excitedly tries a freshly baked simit, then laughs and gives the camera an approving reaction.
She continues toward the Grand Bazaar, where she browses colorful handmade ceramics and shimmering Turkish lamps. She picks up a tiny mosaic lamp, holds it beside her face, and playfully asks the camera, “Should I take this home?”
Cut to her sitting at a rooftop café overlooking the Bosphorus at sunset, sipping Turkish tea while ferries glide across the water and seagulls pass overhead. She captures the view, then turns the phone back toward herself with the golden sunset behind her.
For the final moment, she walks along the waterfront as Istanbul's evening lights begin to glow. She smiles directly into the camera and says, “I think I’m already falling in love with Istanbul. See you on the next adventure!”
Authentic handheld smartphone footage, natural selfie-camera perspective, subtle camera shake, realistic autofocus and exposure shifts, genuine facial expressions, natural walking and hand movements, realistic skin texture, detailed fabric and jewelry, believable crowds and background activity. Authentic Istanbul ambience with ferry horns, distant conversations, footsteps, seagulls, street vendors and café sounds. Warm natural lighting, realistic sunset, cinematic but not overly polished. No beauty filter, no artificial skin, no exaggerated acting, no CGI appearance, no distorted hands, no text overlays. Maintain the exact same woman, outfit, hairstyle, facial features and proportions throughout the entire video.
CHARACTER CONTINUITYで外見を固定してから街を急ぐ場面を並べ、現代の車や建物を除外します。
プロンプトを見る・コピー
原文プロンプト
プロンプト全文 · 3208文字
Create a cinematic, ultra-realistic 1980s lifestyle story about a stylish young woman who has one goal: make it to a legendary rooftop dance party before the final song begins. The entire 30-second video follows her race through the city, building anticipation until she finally reaches the party.
CHARACTER CONTINUITY
The same young woman in every scene, identical face, hairstyle, body, makeup, and clothing.
She has voluminous wavy dark-brown hair, subtle 1980s makeup, gold hoop earrings, and a confident but expressive personality.
Exact same outfit throughout: oversized cream blazer with rolled sleeves, fitted black top, high-waisted light-wash jeans, white sneakers, black shoulder bag, and sunglasses.
SCENE 1 — THE GOAL | 0–7 SEC
Morning in a vibrant 1980s city. She receives an urgent handwritten note from her best friend: “ROOFTOP PARTY — 8 PM. DON’T BE LATE!”
She looks at the clock, realizes she is running out of time, grabs her shoulder bag, puts on her sunglasses, and rushes outside.
Quick cinematic close-up of the handwritten invitation, clock, her determined expression, and her sneakers hitting the pavement.
SCENE 2 — THE RACE | 7–14 SEC
She jumps into a classic 1980s car and races through the colorful city streets.
She checks the clock on the dashboard and realizes the party is getting closer to starting. She looks worried but determined.
Fast cuts: vintage traffic, neon signs, passing storefronts, her hands on the steering wheel, wind blowing through her hair, and the city rushing past the windows.
SCENE 3 — LAST CHANCE | 14–22 SEC
She arrives near the party but discovers the elevator is broken.
She grabs her bag and runs up a long staircase, climbing several flights toward the rooftop. Her friends are already waiting above, looking toward the door.
She reaches the final staircase, exhausted, then hears the distant sound of music starting.
She smiles and pushes herself harder.
SCENE 4 — THE PAYOFF | 22–30 SEC
The rooftop doors suddenly swing open.
She arrives just as the first beat of the final song begins. Her friends cheer excitedly and pull her into the party.
City skyline glowing behind them, colorful 1980s lights, dancing crowd, balloons, vintage speakers and neon decorations.
She laughs, removes her sunglasses, and joins the dance.
FINAL SHOT: The camera pulls backward as she dances confidently with her friends against the glowing city skyline. She looks directly into the camera with a triumphant smile.
VISUAL STYLE
Authentic 1980s cinematic lifestyle aesthetic, nostalgic 35mm film texture, subtle VHS imperfections, soft halation, practical neon lighting, realistic skin and hair, period-correct cars, architecture and props, energetic music-video editing, smooth tracking shots mixed with handheld movement, dramatic close-ups, natural expressions, premium fashion-commercial quality.
CRITICAL CONTINUITY: Same woman, identical face, identical hairstyle, identical outfit and accessories in all four scenes. The story must clearly communicate goal → urgency → obstacle → achievement.
No smartphones, modern vehicles, modern clothing, modern buildings, text overlays, subtitles, logos, watermarks, or futuristic objects.
STYLE REFERENCE: Photoreal, vibrant open-world action-game aesthetic — naturalistic sun-drenched color grading, dynamic third-person gameplay camera, cinematic framing typical of open-world exploration moments DURATION: 10 seconds, single continuous take, no cuts ASPECT: 9:16 vertical SETTING: A sunny tropical beach at midday — turquoise water, white sand, scattered palm trees, gentle rolling waves, bright natural sunlight with soft lens flare off the water CHARACTERS: - MAIN CHARACTER:
@image1
— use this reference image as the exact locked visual identity, maintaining the same face, hair, body proportions, and outfit precisely as shown, from every angle, with no design deviation - WOMAN: a beautiful woman standing near the shoreline in casual beachwear, smiling warmly — generic, non-specific appearance CAMERA: Third-person perspective, open-world-game-style follow-cam. Behind-the-shoulder tracking during the walk and run, then a dynamic arcing camera move that swings around the character during the single backflip to capture it cinematically, settling low near the water surface for the dive and splash. HUD OVERLAY: A small minimap in the bottom-left corner of the screen (radar-style circle showing the character's position), unobtrusive — no other UI elements. BEAT-BY-BEAT: 0:00–0:02 — Protagonist walks onto the beach, camera trailing behind the shoulder; they spot the woman standing near the waterline 0:02–0:03 — Protagonist raises a hand and waves, calling out "Hello!" with a warm grin; the woman smiles and waves back 0:03–0:06 — Protagonist turns and breaks into a run down the beach toward the water, sand kicking up behind them, camera dynamically following at running speed 0:06–0:08 — Reaching the shoreline, the protagonist leaps upward and performs exactly one single backflip — one complete 360° rotation, performed only once — as the camera arcs fluidly around the rotation to capture it from a dramatic angle, sunlight flaring across the frame mid-rotation 0:08–0:10 — Protagonist completes the single rotation and plunges into the water with a splash; camera settles low at the water's surface, capturing the splash and rippling water sparkling in the sunlight CHARACTER DIALOGUE: "Hello!" (warm, friendly, called out while waving) — no on-screen subtitles beyond the minimap HUD AUDIO: Gentle ocean waves, seagulls calling, a light breeze, bare feet on sand, the friendly greeting line, a satisfying splash as the character hits the water, an upbeat breezy ambient undertone NEGATIVE PROMPT: no on-screen text or subtitles, no logos or brand names, no extra HUD elements beyond the minimap, no watermark, no extra unintended cuts, no extra limbs, no design drift on the protagonist from the reference image, no first-person view, no sexualized content, no real identifiable person depicted as the woman, no multiple flips or repeated rotations — only one single backflip performed once
Cinematic fashion commercial, 30 seconds, New York City streets at golden hour. A young woman with short platinum blonde hair, fair skin, walking confidently and stylishly through busy NYC crosswalks and sidewalks. She wears a shiny metallic silver cropped jacket with high collar, matching mini skirt, and knee-high silver metallic boots. She carries a structured red leather Prada handbag with the triangular Prada logo clearly visible.
Camera shots sequence:
- Low-angle close-up of her silver boots stepping on white crosswalk lines
- Medium shot of her walking forward, holding the red bag, slight smile
- Close-up beauty shots of the red Prada bag with logo and gold hardware, warm sunlight flare
- Wide cinematic shot of her walking down a sunny NYC avenue with yellow taxis, tall buildings, scaffolding, and traffic lights
- Overhead shot of her crossing the street
- She walks past luxury store windows
- Low-angle dramatic shot of her holding the bag with the text “SET THE PACE” overlay
- Side profile walking past yellow taxis
- Final confident pose in the middle of the street with taxis and people in background
- Ends with black screen and white text: “NEW YORK OWN THE STREETS”
Style: High-fashion luxury advertisement, glossy metallic textures, warm golden sunlight, cinematic color grading, shallow depth of field on bag close-ups, dynamic yet elegant movement, premium Prada campaign aesthetic. Realistic, high detail, 4K.
Create a 17-second cinematic photorealistic video using the provided character reference and the provided master storyboard as strict visual references. Recreate the storyboard sequence accurately without changing the characters, environment, props, wardrobe, or story.
Shot 1 — 0.0–1.3s: A woman stands inside an office/building and carefully aims a large telescope toward a distant building. Cinematic medium shot, subtle camera movement, realistic lighting.
Shot 2 — 1.3–2.5s: Extreme close-up of her eye looking through the telescope. Detailed eyelashes, realistic skin texture, focused expression, shallow depth of field.
Shot 3 — 2.5–3.8s: Close-up of the telescope as she adjusts and aims it precisely. Smooth cinematic camera movement, realistic metal and glass reflections.
Shot 4 — 3.8–5.0s: Telescope point-of-view shot showing the distant building through the lens. Clear optical perspective, realistic depth and slight lens distortion.
Shot 5 — 5.0–6.3s: Exterior/front view of the telescope, maintaining the same woman and location. Build anticipation with a subtle push-in.
Shot 6 — 6.3–7.5s: A piece of fried chicken suddenly enters the frame from the side, creating a humorous surprise. Natural motion and realistic physics.
Shot 7 — 7.5–8.8s: Cut to the group inside the office near the window, watching the situation with surprised expressions. Keep their appearance and environment consistent with the storyboard.
Shot 8 — 8.8–10.0s: The fried chicken flies rapidly across the scene toward the woman. Use dynamic tracking camera movement and realistic motion blur.
Shot 9 — 10.0–11.4s: The woman catches/bites the fried chicken naturally. Close cinematic framing, realistic facial expression, comedic timing.
Shot 10 — 11.4–13.0s: She calmly returns to the telescope and continues looking through it as if nothing unusual happened. Smooth transition, deadpan comedy.
Shot 11 — 13.0–14.2s: Cut back to the group; they begin eating fried chicken together.
Shot 12 — 14.2–15.8s: The group continues eating happily. Natural gestures, subtle reactions, realistic food movement.
Shot 13 — 15.8–17.0s: Final cinematic group shot of everyone eating together. Hold for a brief comedic ending.
Overall style: ultra-realistic cinematic commercial, natural human movement, realistic physics, consistent faces and clothing, detailed food textures, realistic office environment, cinematic depth of field, subtle handheld camera movement, smooth cuts, natural motion blur, believable lighting and shadows, high-end film production quality, 4K photorealism.
Important: Follow the provided storyboard exactly. Do not invent new scenes, do not change the sequence, do not change the characters, and do not add unnecessary objects or actions.
Heartwarming Pixar-style 3D animated scene, 10 seconds, vertical 9:16. A child's cozy bedroom at night, moonlight through the curtains, toys scattered on the rug. Under the bed hides a big fluffy purple monster with round glowing eyes, tiny horns, floppy ears, and soft teddy-bear fur, trembling and hugging a small stuffed bunny.
0–3s: Slow push-in under the bed. The monster peeks out nervously as the hallway light clicks off, plunging the room into darkness. It gasps and pulls its blanket over its head, eyes wide with fear.
3–6s: Creaky floorboards and a ticking clock. A shadow from the closet makes the monster squeak and shiver, its fur puffing up. It whimpers softly.
6–8s: A small hand reaches down from the bed holding a glowing star-shaped night-light. The monster blinks, then slowly smiles as warm golden light fills the space beneath the bed.
8–10s: The monster happily cuddles its bunny and the night-light, sighing in relief, and gives a sleepy yawn as the child's hand pats its fuzzy head. Twinkling music box melody.
Soft cinematic lighting, cozy and funny, expressive character animation, smooth motion, gentle sound effects and soothing lullaby score.
Photorealistic cinematic live action, night photography, low-key lighting, deep teal and black water, warm practical lantern glow, film grain.
One old fisherman sits alone in one small wooden rowboat on a perfectly still black sea under a starless sky. One oil lantern hangs from the boat.
0-5s: The fisherman sits in silence, holding a line in the water. The sea is flat like glass, reflecting the lantern. The viewer is wide and low across the water. Sound: total quiet, a faint creak of wood.
5-10s: A deep church bell rings from somewhere far below the water. Ripples spread out from beneath the boat. The fisherman slowly leans over the edge and looks down. The viewer looks down past him into the water, where hundreds of faint golden lights glow in the depths.
10-15s: An entire sunken city rises around the boat, stone towers and rooftops breaking the surface on all sides with water pouring off them, every window lit with lanterns. The boat now floats in a flooded city street. The fisherman stands, staring up at the glowing bell tower directly in front of him as the bell rings again. Sound: the bell, rushing water, then stillness.
Negative prompt: cartoon, 3D animation, stylized, CGI look, extra people, extra boats, distorted hands, text, watermark, slow motion, flickering, warped buildings
Style: cinematic magical realism, soft overcast daylight, rich color, 35mm film look, shallow depth of field.
0-5s: One woman in paint-stained overalls paints a bright red wooden door onto a plain brick wall in an empty alley, the brush strokes thick and wet, the viewer behind her shoulder.
5-10s: She finishes the brass doorknob with a final stroke, sets down her brush, reaches out, and the painted knob turns in her hand, the painted door swings open into warm golden light, the viewer following her as she steps through.
10-15s: She steps out of a large framed painting hanging on the wall of a quiet art museum, landing on the polished floor, two visitors holding brochures freeze and stare at her, she nods politely and walks past them.
Negative prompt: more than one woman, more than two visitors, duplicate people, cartoon style, graffiti, text, watermark, slow motion
A dramatic 30-second cinematic parkour chase following one elite male athlete descending through an ancient Italian seaside town built vertically into a steep mountain. Every cut places him farther downhill toward the sea. Keep the same bell tower, roofs, cliffs and harbor visible from progressively lower elevations. Photorealistic feature-film cinematography, anamorphic lens character, natural motion blur, subtle grain and rich Mediterranean color.
Consistent athlete: handsome Mediterranean man, late 20s, lean powerful build, olive complexion, short dark curls, light stubble. Faded charcoal athletic T-shirt, rust-red training trousers, black wrist wraps and dark parkour shoes. A world-class competitive parkour athlete: explosive, fearless, fast and technically precise. Never looks at camera.
0.0–3.5s: Epic drone shot above an old Italian town clinging to the mountain. Ochre, rose, cream and faded-blue buildings plunge toward a fishing harbor and cobalt sea. Terracotta roofs, stone stairs, arches, balconies, cypress and bougainvillea form a vertical maze. Dark clouds crown the mountain while sunlight breaks across the water. The athlete stands on a high roof, studies the route, then launches into a sprint as the drone dives after him.
3.5–7.8s: First hero stunt, uninterrupted wide side-tracking shot, entire body visible. He accelerates across a sloping roof and commits to a long roof-to-roof precision jump over a broad open alley. Show the complete takeoff, airborne distance and landing. He lands on the narrow edge of the lower roof, absorbs the force, regains speed and continues downhill. The sea flashes far below.
7.8–12.0s: Second hero stunt, one continuous descending cat leap. Camera tracks beside and below him. He sprints off the roof toward a building several feet lower, flies across the street, catches a wrought-iron balcony with both hands and plants both feet against the wall. Without pausing, he pulls himself over the railing and vaults onto a lower terrace. Show full body, clear height difference, powerful grip and convincing momentum.
12.0–16.5s: Third hero stunt across a multi-level market square. A stabilized low camera races parallel. He bursts through an arch, accelerates toward a broad stone fountain and performs a long diving kong vault over its full width. His legs pass cleanly between his arms before he launches beyond the fountain into a substantial drop to the lower plaza. He lands in a fast shoulder roll, rises immediately and races onward as pigeons explode into the air.
16.5–21.0s: Fourth hero stunt down a deep exterior stair canyon. Wide crane view keeps his whole body visible. He runs along a high stone railing, jumps to the opposite wall, plants one foot high against the stucco, pushes away and clears the stairwell to a landing several levels below. He lands on both feet, converts the impact into a roll and continues toward the harbor. The camera drops with him, revealing the sea filling the background.
21.0–25.7s: Cliffside finale. Waves crash against black rocks below. In one tracking shot, he races down the last cliff path, vaults a chest-high stone barrier and launches into a dramatic running front flip over the descending harbor steps. Show the full rotation in profile, his body isolated against the silver-blue sea, then a firm landing on the lower promenade. He runs out of it as sunlight breaks beneath the clouds and turns the coast gold.
25.7–30.0s: The speed gives way to intimacy. Ground-level medium-wide beside the sea. He slows and stops at the water’s edge, breathing hard. A lovely Mediterranean woman in her late 20s waits near the surf, with wavy dark-brown hair, pale-yellow summer dress and flat sandals. She hears him, turns naturally and gives him a warm smile. His intensity softens into a relieved smile. He walks toward her while wind moves her hair and dress. End with a slow pull-back as they meet beneath the town rising above them.
Camera: fast stabilized pursuit, wide side tracking, descending crane moves and sweeping coastal reveals. Never crop him during a stunt or cut between takeoff and landing. Keep him fully visible during every major move. Preserve constant downhill direction and make the sea larger after every transition.
Color and mood: terracotta, ochre, limestone, Mediterranean blue, storm-grey, white spray, rust red and warm gold. Urgent, athletic, exhilarating, romantic and triumphant, like a major action-film climax.
Maintain his exact face, hair, physique and clothing. One athlete only. No duplicates, teleportation, upward travel, timid hops, ordinary jogging, easy playground vaults, broken anatomy, extra limbs, costume changes, text, logos, cars or crowds. The woman appears only in the final shot. Powerful human gravity and weight, 24 fps cinematic motion, 16:9, exactly 30 seconds.
A cinematic architectural odyssey following a solitary older female archivist through an immense abandoned coastal city. Photorealistic, ultra-high detail, sophisticated filmic color science, anamorphic lens character, subtle film grain, deep atmospheric perspective, rich natural contrast. Match the reference only in pacing, camera language, visual scale, and seamless montage structure. Do not reproduce its landscapes, traveler design, vehicles, or compositions.
Consistent subject: Tall Black woman in her early 60s, lean build, close-cropped natural silver hair. Long burnt-orange weatherproof coat, slate-blue scarf, charcoal trousers, black leather gloves, worn black ankle boots, compact dark-brown archival satchel worn across her body. Calm, purposeful presence. She never looks at the camera and is always studying the abandoned city.
Shot-by-shot timeline with precise pacing and camera language:
0.0–1.8s
Wide static shot transitioning into a very slow push-in. The woman stands alone in the center of a monumental abandoned railway terminal. The vaulted glass roof is partially broken, allowing pale morning light and fine rain to enter. A thin layer of water covers the marble floor, producing a vast mirror reflection. Empty platforms and silent iron arches recede into atmospheric haze. Her burnt-orange coat is the only warm color in the frame. Camera creeps forward and rises slightly, emphasizing the immense scale of the terminal around her.
1.8–3.1s
Low rear tracking shot at medium-wide distance. She walks steadily through a canyon-like avenue between enormous brutalist concrete towers. Wind drives loose paper and rainwater across the pavement. Long elevated walkways disappear into low cloud between the buildings. Her coat moves heavily in the wind while her satchel remains secure. Slight organic camera sway, controlled and deliberate, following several steps behind her.
3.1–4.2s
Locked-off symmetrical wide shot inside a ruined opera house. The woman stands ankle-deep in perfectly still water at the center aisle, facing the distant stage. Rows of faded crimson seats emerge from the reflective water. Torn velvet curtains move almost imperceptibly beneath a collapsed ceiling filled with cold blue daylight. Her reflection remains clear and undisturbed. Quiet, solemn, architectural stillness.
4.2–5.9s
Smooth aerial pull-back with a slight orbit. Begin close behind the woman on the roof of the opera house, then reveal the entire drowned city surrounding her. Concentric avenues have become dark canals. Pale stone towers, domed civic buildings, abandoned bridges, rooftop gardens, and a massive curved seawall extend toward the ocean. Low cloud spills between the tallest structures. The woman becomes a tiny burnt-orange point within the immense geometric city. Floating, stable drone movement.
5.9–7.8s
Night sequence, extreme wide shot. A single vintage crimson tram travels slowly left-to-right across a high steel viaduct above flooded streets. The woman appears only as a small silhouette inside one illuminated window. A silent violet electrical storm flashes behind the skyline, briefly revealing cranes, antennas, distant towers, and rain suspended in the air. The camera remains locked. Reflections of the tram lights stretch across the black water below. No explosions, destruction, or frantic motion.
7.8–9.0s
Golden-hour peak shot. The woman stands on a vast rooftop communications deck surrounded by enormous weathered parabolic antennas. She faces left toward the ocean, one gloved hand resting on her satchel. The antennas form monumental circular silhouettes against glowing amber and rose-colored clouds. Warm rim light outlines her silver hair and orange coat. Slow, subtle push-in. Gentle coastal wind moves her scarf.
9.0–11.6s
Epic aerial tracking shot. A crimson tram moves along a curving elevated railway between monumental skyscrapers and across broad sections of the flooded city. Below, geometric plazas, narrow canals, rooftop forests, empty bridges, and pale stone towers extend toward a brilliant blue sea. The camera flies parallel to the tram and slightly above it, then gradually pulls farther away until the tram becomes a small moving line within the city. Crisp daylight, deep blue sky, scattered sculptural clouds, smooth high-speed drone motion.
11.6–14.0s
Intimate point-of-view from inside the stationary tram. The edge of a dark window frame and softly reflected interior lights remain out of focus in the foreground. Outside, the same woman stands alone on an elevated platform, her burnt-orange coat, slate-blue scarf, brown satchel, silver hair, and black gloves unchanged. She holds a large translucent architectural map against the light and studies the distant flooded skyline through it. The city remains sharp beyond her, with white towers, blue water, and rotating antennas under late-afternoon light. A gentle breeze moves the map and her scarf. She never turns toward the camera.
Overall cinematography and technical direction:
Camera language: contemplative locked wides, slow creeping push-ins, controlled rear tracking, seamless floating aerial reveals, and one intimate vehicle-interior point of view. Stable, elegant motion throughout. No fast cutting, whip pans, shaky footage, abrupt zooms, or direct-to-camera performance.
Lighting progression: pale rainy morning, severe white overcast, cold blue interior light, violet electrical night, amber golden hour, crisp coastal daylight, soft late-afternoon light.
Color science: steel blue, wet charcoal, oxidized copper, faded crimson, and pale stone, unified by the woman’s burnt-orange coat. Subtle anamorphic flare, realistic reflections, restrained halation, fine cinematic grain, natural skin texture, volumetric weather, and deep atmospheric separation.
Mood: quiet discovery, lost human grandeur, the scale of architecture against one small observer, melancholy without horror.
Maintain exact character, wardrobe, coat color, satchel, age, body proportions, and facial identity in every appearance. No additional people. No readable text, logos, subtitles, signs, fantasy creatures, modern traffic, or post-apocalyptic debris piles. 24 fps cinematic motion, realistic physics, 4K detail, 16:9 widescreen.
30-second cinematic high-speed mecha flight sequence through a dense futuristic megacity.
A single sleek advanced humanoid mecha in compact high-speed flight configuration, dark metallic armor with subtle white panels, small glowing blue thrusters, aerodynamic silhouette, consistent design throughout the entire video. No other mecha, no combat, no weapons firing.
The scene takes place in a gigantic futuristic city at late afternoon, with extremely tall skyscrapers, glass curtain walls, steel structures, elevated highways, rooftop machinery, signs, antennas and narrow urban canyons. Realistic large-scale architecture, cinematic lighting, physically believable reflections and motion blur.
The main goal is extreme sensation of speed, dangerous proximity to architecture, immersive camera movement, strong parallax, and realistic inertia.
Use clear hard cuts every 3 seconds while preserving the same mecha, same city, same lighting, and continuous direction of travel.
[00:00-00:03]
24mm ultra-wide lens.
Low rear chase camera, flying only several meters behind the mecha.
The mecha accelerates violently through a narrow canyon between skyscrapers.
Building walls are extremely close on both sides.
Windows, beams, pipes and signs rush past the edges of frame at enormous speed.
Strong foreground parallax.
Camera vibration from turbulent air, subtle realistic handheld instability.
The mecha remains sharp while the nearby architecture streaks with motion blur.
[00:03-00:06]
Hard cut.
Side tracking shot, extremely close to the right side of the mecha.
Camera flies parallel to the mecha only a few meters away.
The mecha skims along a huge glass skyscraper wall with less than one meter of visual clearance.
Rows of windows flash past like rapid horizontal stripes.
The camera almost brushes the wall.
Extreme sensation of dangerous proximity and speed.
[00:06-00:09]
Hard cut.
Front three-quarter tracking shot, camera flying backward ahead of the mecha.
18mm wide-angle lens.
The mecha rapidly approaches the camera while weaving between structural beams connecting two buildings.
Camera struggles to maintain distance and slightly drifts from inertia.
Foreground beams suddenly whip past the lens.
The city behind the mecha compresses into a dense high-speed tunnel.
[00:09-00:12]
Hard cut.
High overhead camera looking almost vertically downward.
The mecha suddenly dives between two enormous skyscrapers.
Camera follows the dive with a slight delay, then aggressively tilts downward.
The city streets far below become visible.
Extreme vertical depth, strong scale contrast, rapid acceleration.
[00:12-00:15]
Hard cut.
Rear chase camera returns.
The mecha approaches the corner of a massive skyscraper at seemingly impossible speed.
At the final moment it banks nearly 90 degrees and turns tightly around the building corner.
The camera does not rotate perfectly with the mecha; it reacts about a fraction of a second late, rolls violently, overshoots slightly, then corrects.
The building wall fills most of the frame during the turn.
Very strong near-miss feeling.
[00:15-00:18]
Hard cut.
Camera mounted extremely low and slightly behind the mecha.
The mecha races beneath an elevated highway and between gigantic concrete support columns.
Columns repeatedly flash past the camera only meters away.
Strong rhythmic foreground occlusion.
The camera shakes from high-speed air pressure and narrowly avoids the structures.
[00:18-00:21]
Hard cut.
Static cinematic camera positioned beside a skyscraper facade.
For the first moment the city appears almost calm.
Suddenly the mecha blasts past the camera from background to extreme foreground at tremendous speed, passing only a few meters from the lens.
Violent motion blur and air distortion during the flyby.
The camera rapidly whip-pans to follow it after it has already passed.
[00:21-00:24]
Hard cut.
Long-lens 85mm tracking shot from another rooftop.
The mecha threads through a dense sequence of skyscraper gaps at extreme speed.
Because of the compressed perspective, buildings appear stacked dangerously close together.
The camera pans rapidly, momentarily losing the mecha behind foreground buildings, then catching it again.
[00:24-00:27]
Hard cut.
Ultra-wide chase shot again, now even closer to the mecha.
The mecha enters an impossibly narrow corridor between two glass towers.
Both walls occupy the left and right edges of frame.
Reflections of the mecha rapidly slide across the glass.
The camera stays directly behind it while the entire environment becomes a tunnel of streaking architecture.
Maximum visual speed.
[00:27-00:30]
Hard cut.
The mecha suddenly exits the narrow skyscraper canyon into a gigantic open section of the city.
The camera follows behind and then rapidly pulls farther back.
The claustrophobic close architecture disappears, revealing an enormous futuristic skyline stretching to the horizon.
The mecha continues accelerating forward into the open city.
Strong contrast between the previous confined high-speed flight and the sudden huge open space.
End on a dramatic wide cinematic composition.
Visual direction:
photorealistic cinematic CGI, premium science-fiction film quality, realistic materials, realistic atmospheric perspective, physically believable high-speed motion, strong foreground parallax, natural motion blur, subtle lens distortion, volumetric sunlight, realistic glass reflections, enormous architectural scale, detailed futuristic megacity.
Camera behavior:
the camera should feel like a physical flying camera with mass and inertia.
It should not perfectly lock onto the mecha.
During rapid turns, the camera reacts slightly late, overshoots, rolls, shakes and corrects naturally.
Nearby objects should pass extremely close to the lens to emphasize speed.
Maintain strict temporal consistency:
same mecha design in every shot,
same armor proportions,
same thruster locations,
same lighting conditions,
same city architecture style,
same direction of travel,
correct scale between mecha and buildings.
Avoid:
slow motion,
floating or weightless movement,
gentle cruising,
perfectly stabilized camera,
static centered framing,
camera clipping through buildings,
mecha clipping through architecture,
changing mecha design,
extra limbs,
deformed geometry,
random transformation,
additional robots,
combat,
explosions,
weapons fire,
unmotivated camera teleportation,
tiny-looking buildings,
miniature city appearance,
cartoon physics.
30 second TV commercial for BMX Juice. 16:9, 1080p, 24fps, live-action cinematic comedy, bright sunny suburban cul de sac. Star: a heavyset 40 year old man with a sweaty mustache and beer belly, wearing a too-tight neon green spandex bodysuit, orange fanny pack, tiny red helmet, one pink knee pad.
0–5s: Drone shot: high aerial over the street, then fast dive down to the man. He straddles the tiny BMX bike, sniffs a bottle of BMX Juice, looks at camera.
5–10s: Low angle tracking shot. He chugs BMX Juice, throws the bottle, pedals furiously. Bike squeaks. He hits a plywood ramp.
10–15s: Drone orbit shot, mid-air. Slow motion 120fps. His face goes from confident to panic. Spandex rips. Juice droplets float.
15–22s: Extreme slow motion. The bike separates from him. He flails like a starfish. A pigeon flies past, also slow motion, looking disappointed.
22–27s: Real-time crash into a giant pile of leaves. Leaves explode. He sits up, dazed, leaf on his face, holds up a fresh BMX Juice bottle.
27–30s: Product shot: BMX Juice bottle on the tiny BMX bike. Text overlay: "BMX JUICE." Tagline: "Get Juiced. Get Wrecked." Audio: funny ukulele, whoosh, squeak, slow-mo rip, thud, upbeat jingle.
Negative prompt: no watermark, no distorted face, no extra limbs, no cartoon, no low quality, no subtitles unless specified.
A 15-second high-energy slapstick comedy video with a handheld smartphone aesthetic, 9:16 vertical frame.
[0-4s] In a cozy living room, AhKit sits on the sofa rubbing his sore neck. Mia stands behind him, smiling angelically at the camera while playfully rubbing her small fists together. Subtitle: "When she offers a massage..."
[4-9s] Snap zoom. Mia cheerfully taps her small fist onto AhKit's shoulder. Sudden dynamic shockwave impact: AhKit's eyes bulge out in dramatic exaggerated horror, his body recoils violently, and he collapses off the sofa onto the floor in hilarious defeat. Subtitle: "CRITICAL HIT! -9999"
[9-15s] Low-angle tracking shot. Mia's sweet face turns into a sinister, villainous mischievous smirk. She grabs AhKit by one ankle with one hand and effortlessly drags him across the wooden floor toward a bedroom doorway. AhKit lies face down, frantically clawing the wooden floor with both hands like a terrified cat refusing to go into the room, legs sliding helplessly. Subtitle: "Ah Kit: Help... ...!"
<<<Image1>>> は Girl A の顔、黒髪の肩上ボブ、青い瞳、体格、制服デザインを定義する。Girl A は標準語を話す真面目な吹奏楽部員で、担当楽器はクラリネット。白背景や元のポーズは継承しない。
<<<Image2>>> は Girl B の顔、茶髪ポニーテール、体格、制服デザインを定義する。Girl B は関西弁を話す明るい吹奏楽部員で、担当楽器はトランペット。白背景や元のポーズは継承しない。
<<<Image3>>> は Girl C の顔、ダークブラウンのボブヘア、体格、カーディガンを含む制服デザインを定義する。Girl C は博多弁を話すおっとりした吹奏楽部員で、担当楽器はフルート。白背景や元のポーズは継承しない。
この3枚の参照画像は3人の別々の女子生徒を定義する。主要人物はこの3人のみ。顔、髪型、衣装、担当楽器、方言、性格を混同しない。
夏休みの高校の吹奏楽部を描く、爽やかで自然な日常アニメ映像。舞台は夏の朝の音楽室。窓の外は強い日差し、青空、校庭の緑。室内には譜面台、椅子、楽器ケース、扇風機、開いた窓、少し揺れるカーテン。外から蝉の声が聞こえる。空気は少し暑く、夏休みの学校らしい静けさがある。字幕なし、ナレーションなし。
全体のカメラ方針：
自然なマルチカット構成。1カットごとに目的を持たせる。引きの状況説明、寄りの仕草、手元の演奏、表情のリアクションを交互に見せる。人物を棒立ちにしないため、会話中も視線移動、重心移動、髪の揺れ、汗をぬぐう、譜面を見る、楽器を持ち直す、息を吸う、指を動かすなどの小さな動作を常に入れる。カメラは不自然に激しく動かさず、ゆるいドリーイン、軽いパン、短い横移動、自然なカットつなぎで構成する。演奏中は表情、指、口元、楽器、視線のやり取りを細かく切り替える。
全体の音：
蝉の声、教室の環境音、椅子を引く音、楽器ケースを置く音、扇風機の回転音、カーテンが揺れる気配、楽器を持つ音、短いチューニング音、息づかい、そして短い合奏。
0〜5秒 — 登場と状況説明
カット1：
音楽室の中を外の窓側から見たやや広めの引き。カメラはゆっくり前進する。朝の光が斜めに差し込み、窓辺のカーテンが揺れている。Girl A は譜面台の前で譜面を見ている。Girl B はドア側から入ってきて、少し前かがみで暑そうにしている。Girl C は楽器ケースを机に置く。
カット2：
Girl A の胸上のミディアムショット。譜面から目を上げ、2人を見る。クラリネットの部品を持ちながら落ち着いて言う。
Girl A が標準語で言う：
{おはよう。最初のフレーズ、合わせよう。}
終了状態：3人ともそれぞれの準備に入る空気になる。
5〜10秒 — 暑さと夏休み感
カット3：
Girl B のミディアムショット。カメラは少し右から寄る。Girl B は制服の襟元を軽くつまみ、片手であおぐようにしながら、汗をぬぐって言う。ポニーテールが少し揺れる。
Girl B が関西弁で言う：
{音楽室あっつ！}
カット4：
扇風機の回転を一瞬映すインサート。続いてカーテンが揺れるカット。夏の空気を見せる。
カット5：
Girl C のやや寄りのミディアムショット。Girl C は楽器ケースを開けながらふっと笑い、フルートを取り出して言う。
Girl C が博多弁で言う：
{でも夏休みの部活って感じやね。}
終了状態：3人の気分がそろい、練習モードに入る。
10〜15秒 — 準備と合図
カット6：
手元の短いモンタージュ。テンポよく3つの短いカットをつなぐ。
- Girl A がクラリネットを組み立てる指先の寄り
- Girl B がトランペットのバルブを軽く押して感触を確かめる寄り
- Girl C がフルートを水平に持ち、キーに指を置く寄り
カット7：
3人の半身が入る横並びのミディアム3ショット。カメラはゆっくり左から右へ小さくスライドする。3人が譜面台に向き直る。Girl A は譜面を見てから2人を見る。Girl B は姿勢を正し、Girl C は小さくうなずく。
カット8：
Girl A の顔寄り。目線で合図し、小さく息を吸って言う。
Girl A が標準語で言う：
{いくよ。}
終了状態：3人が同時に吹き始める直前。
15〜24秒 — 合奏メイン
演奏パートは会話よりも長く取り、部活らしい集中を見せる。カメラはテンポよく切り替えるが、空間の向きは崩さない。
カット9：
3人を正面から捉えたミディアム3ショット。合図と同時に演奏開始。最初の1〜2秒はこのカットで、3人が同時に入る瞬間を見せる。息を吸ってから音を出す流れを自然に見せる。
カット10：
Girl A のクラリネットの手元と口元の寄り。指が動き、真剣な視線が譜面へ向く。
カット11：
Girl B のトランペットのややローアングル寄り。ベルの輝き、頬の張り、バルブを押す指の動き、リズムに少し乗る体の揺れを見せる。
カット12：
Girl C のフルートの横顔寄り。フルートの水平ライン、指先、落ち着いた息遣い、やわらかな目線を見せる。
カット13：
再び3人の斜め前からの3ショット。カメラがわずかにドリーインし、3人の音がまとまっていく感じを見せる。演奏中、Girl A は譜面と2人を交互に見て、Girl B は少し楽しそうに吹き、Girl C は安定して支える。
カット14：
短いリアクション挿入。Girl A の目線、Girl B の口元、Girl C の指先をリズムよく短く切り替えて、演奏の躍動感を出す。
終了状態：フレーズを吹き終え、音が止まり、3人が手応えを感じる。
24〜27秒 — リアクション
カット15：
演奏直後の3ショット。音が止まり、一瞬だけ静けさ。Girl B が顔を上げて、嬉しそうに笑う。Girl A は少しほっとし、Girl C はうなずく。
カット16：
Girl B の寄り。トランペットを少し下ろして、明るく言う。
Girl B が関西弁で言う：
{今のええやん！}
終了状態：3人が「もう一回いける」という前向きな空気になる。
27〜30秒 — 締めの再演奏
カット17：
Girl C のミディアム寄り。フルートを構え直し、笑顔で短く言う。Girl A が少し笑い、Girl B もすぐ構え直す。
Girl C が博多弁で言う：
{このままいけるばい。}
カット18：
すぐに3人の正面ミディアム3ショットへカット。最後の2秒ほど、もう一度だけ短いフレーズを揃えて演奏する。今度は最初より自然に息が合っている。カメラはごくゆるく前進する。
カット19：
締めの短い寄り。演奏を終えたあと、3人が顔を見合わせて笑う。夏の光が差し込む音楽室で、前向きな空気のまま終わる。
全編の連続性ルール：
Girl A は常に標準語を話し、常にクラリネット担当。
Girl B は常に関西弁を話し、常にトランペット担当。
Girl C は常に博多弁を話し、常にフルート担当。
各キャラクターの顔、髪型、衣装、性格、方言、担当楽器を全編で一定に保つ。
主要人物は3人のみ。追加の部員を主役として出さない。
夏の朝の音楽室、照明、暑さ、学校の雰囲気を全編で維持する。
会話中は口だけでなく、視線、肩、手、重心移動、髪の揺れなども自然に動かす。
演奏中は息を吸う、指を押さえる、楽器を支える、体が少し揺れるなど、演奏に伴う自然な身体動作を入れる。
字幕、テロップ、ナレーション、ファンタジー演出、衣装変更、キャラ入れ替わりは禁止。
[Goal]
Morphing from <<<Image1>>> to a werebunny (a anthropomorphic bunny) keeping background .
[Prompt]
The woman seems in great joy, she shakes her head smiling and seductive, her face starts to stretch like bubbling, she makes funny sounds and we see little front tooths. A muzzle starts to stretch as she continues to move left and right. woman transmorphs into the werebunny, 1990 pratical effects, bunny ears growing , white fur growing , teeths becoming like bunny teeths : metamorphing into werebunny. Her eyes start glowing red as she transforms her trasformation then suddently turn into kawai eyes. The werebunny cleans her muzzle with her two paws .
[Audio Style]
live sounds include organic transformation, smoothe funny sounds .
[Camera & Core Constraints]
Continuous shot filmed hand held with a sony handicam. Camera corrections, focus hunting and minizoom ins and outs, subject always facing camera. Some chromatic artefacts on edges. no magical appear and no alpha blending. do not change the identity of the characters throughout the entire video. Make the transitions smooth and logically plausible.
Generate a single continuous shot focused on natural human performance. Nearly empty commuter train at 2:00 a.m., moving through a rain-soaked city. A tired Black woman in her early forties sits opposite an older Asian man holding a familiar red umbrella. Reflections from passing streetlights move realistically across the windows and their faces. The camera begins in a wide symmetrical composition, then slowly dollies toward the woman as she recognizes the umbrella. Her expression changes subtly from fatigue to uncertainty and finally quiet recognition. Without looking directly at him, she asks softly, “Where did you get that umbrella?” The man studies her for a moment and replies, “You gave it to me twenty years ago.” Hold on her restrained reaction rather than exaggerated shock. Grounded contemporary drama, realistic commuter lighting, wet glass, muted blue-gray palette with red as the only strong accent, natural micro-expressions and body language. Audio: train wheels, rain against glass, distant station announcements without intelligible words, quiet American English dialogue. No music.
Generate a 30-second, 16:9 animated short: a young woman searches a hill town for a landmark with a paper map.
ART STYLE — hold this exactly in every frame, it is the contract: painterly 3D concept art. Sculpted readable silhouette. Planar cheek structure and wedge-shaped hands. Directional hand-painted brush patches locked to the 3D surfaces, never floating. Rough matte cloth. Ivory light anchor against deep oxblood architecture, with cobalt accents. Broad morning key with a restrained rim. 50mm eye-level framing. Layered staging: foreground, character plane, background. NO plastic skin, NO smooth airbrush, NO floating brush texture, NO photorealism, NO gradients, no text, no logos, no watermarks, no extra people beyond those described.
CHARACTER: @Image1 is the woman and the style authority. One original adult courier — long asymmetrical face, cropped silver-black bob, ivory ankle-length coat with oxblood piping, cobalt satchel. Keep her face, hair, coat and satchel identical in every frame. Her expression arc runs from breezy confidence, through mounting frustration, to defeat, to startled indignation, to open joy.
Shot 1 [0-5s]: THE MAP. She stands in a hill-town tram plaza unfolding a paper map far too big for her arms, kiosk stone steps and one wind-bent tree behind her. She turns the map ninety degrees, then upside down.
Shot 2 [5-11s]: THE FIRST GUESS. She strides off down a side street, map held out in front of her like a shield, counting buildings under her breath — and arrives at a blank stone wall. She looks up at it, then back at the map.
Shot 3 [11-17s]: THE SECOND GUESS. She rotates the map and sets off the other way, weaving through market stalls and a fountain square, stepping over a cat, checking the map against a street sign that plainly does not match.
Shot 4 [17-21s]: GIVING UP. She stops dead in the middle of the plaza. The enormous map sags in her hands. Her shoulders drop. She blows a strand of hair off her face, beaten.
Shot 5 [21-25s]: THE BUMP. A passer-by clips her shoulder hard as they hurry past. She staggers half a step, snaps her head round after them, indignant, and calls out sharply: "Hey!" — young female voice, irritated, clipped.
Shot 6 [25-30s]: THE REVEAL. Mid-turn she stops dead — because the landmark she has been hunting is standing directly behind her, enormous, filling the frame. Hold close on her face: her eyes blink once, twice, the irritation draining out of them; her mouth opens slightly; and the expression blooms slowly into an unguarded joyful smile. The camera then pulls back in one continuous move to a wide shot — she stands small at the base of the landmark, her back to camera, head tilted up, looking at it. The map hangs forgotten at her side. Hold on the wide.
CAMERA: tracking shots at her shoulder through the searching beats; locked and still for the defeat; a fast whip round with her on the bump; then one continuous accelerating pull-back for the final reveal that never decelerates before the last frame.
AUDIO: light playful woodwind-and-pizzicato score that stumbles with each wrong turn, drops out completely on the bump, and resolves into one warm full phrase as she smiles. Hill-town ambience, paper crinkling, footsteps on stone, market chatter, a tram bell in the distance. The only dialogue in the entire video is her single word: "Hey!".
Create a 30-second cinematic sci-fi short set in a mysterious floating city above a dark, endless ocean during a violent nighttime storm. Build the story around escalating anticipation, visual clues, and a shocking final reveal.0-5 seconds: Begin with an enormous wide shot of the floating city barely visible through thick storm clouds. Lightning briefly illuminates the skyline. The city appears completely abandoned. Rain falls heavily and the ocean below is unnaturally still.5-10 seconds: Slowly descend into the empty city. Flickering lights activate one by one along deserted streets, moving deeper into the city as if something is approaching. A distant mechanical rumble grows louder. No people are visible.10-15 seconds: Follow the lights toward the edge of the city. The camera looks down through the clouds. Far beneath the ocean surface, an enormous circular light suddenly appears. The entire floating city trembles.15-20 seconds: Everything goes completely dark. The storm stops for a moment. The ocean below begins rotating in a gigantic perfect circle. Buildings shake as deep mechanical sounds emerge from beneath the water. Cracks of blue light spread across the ocean surface.20-25 seconds: The camera rapidly pulls backward and upward. Massive structures begin rising from beneath the ocean, initially appearing to be enormous towers. The floating city starts moving upward with them.25-30 seconds: Final reveal. Pull far enough back to reveal that the "floating city" is actually a tiny structure built on the back of a colossal humanoid-like machine that has been lying dormant beneath the ocean for centuries. Its enormous head rises from the darkness, its eyes slowly opening beneath the storm clouds. The city is revealed to be only a small civilization living on its back.End on the creature looking toward the horizon as lightning illuminates its full silhouette.Visual direction: photorealistic cinematic sci-fi, dark storm atmosphere, massive scale, realistic rain and ocean simulation, volumetric clouds, eerie blue illumination, deep shadows, abandoned futuristic architecture, subtle camera shake, slow controlled camera movement that accelerates during the reveal, realistic reflections, cinematic depth of field, high-detail environments, 4K, 24fps.Important: Keep the creature completely hidden until the final 5 seconds. Do not show obvious clues that reveal its true form too early. The audience should initially believe they are watching a floating futuristic city, then realize in the final moments that the entire city has been sitting on something alive and unimaginably enormous.
Cinematic slow forward dolly shot moving past rugged rocks and weathered foliage toward a lone man standing tall on a jagged mountain peak. In the foreground, an weathered wooden suspension rope bridge with worn timber planks sways gently in the breeze, bridging an alpine cliff trail to the mountainside. The camera pushes in past his shoulder, gradually revealing the dizzying scale of the deep gorge below and the rushing, turquoise-blue river cutting through the canyon floor.
——（作者後續回覆）——
The orbital camera movement. It moves around showing everything around him.
Prompt:
Sweeping circular aerial orbit shot smoothly rotating around a solitary man standing at the summit of a dramatic mountain ridge looking down into an abyss. As the camera circles, it reveals the rustic wooden suspension bridge hanging precariously across a deep canyon gap connecting the mountain trail, then sweeps across the vast gorge and the winding white-water river carving through steep rock walls below, set against dramatic golden hour light.
This one is the Tilt-Down / Pedestal Crane Shot which moves down tilting to show what is below.
Prompt:
Vertical crane tilt shot beginning at eye level with a man looking out from an isolated rocky mountain peak. The camera rises and tilts steeply downward, tracing the vertical rock face to reveal a narrow suspended wooden plank bridge spanning an expanse along the trail, continuing down to the bottom of the sheer gorge where a powerful river churns with white water amidst rocky rapids.
Wan 3.0 Prompt
Create a 15-second action-comedy video with synchronized audio. Use the selected aspect ratio. Serious sword-action buildup, one spectacular stroke, an ordinary cake-cutting reveal, then A's familiar stunned face.
REFERENCES
A = the black-haired heroine in Image 1.
B = the white-haired heroine in Image 2.
Preserve their exact faces, hairstyles, costumes, accessories, proportions and original 2D anime art style. Appearance references only; new poses and background. Keep identities and character scale consistent.
A's voice timbre references Audio 1. Use its vocal identity, not its recorded words.
B has no voice reference and remains completely silent, with closed lips.
SETTING AND STYLE
Empty rooftop above Shibuya at dusk. Photorealistic rooftop and distant skyline; hand-drawn 2D anime characters with clean lines and cel shading. Match lighting and contact shadows while preserving anime faces.
A stays screen-left, B screen-right. One sturdy table stands between them toward the foreground. Maintain screen direction.
A holds one katana throughout. One plain round white-frosted cake rests on one wooden cutting board, initially on B's side of the table and visible from the first frame. One fork rests on a ceramic saucer beside B.
B uses A's serious sword technique to cut her cake.
Shot 1 (0.0-4.0 seconds): THE FINISHER
Low-angle medium-wide two-shot showing both characters, table and cake.
A plants her feet and raises the katana overhead with a two-handed grip. Bright teal energy coils around the blade. Wind lifts her hair and costume edges; one circular light ripple spreads beneath her feet. Keep her face, hands and sword clearly visible.
B looks from the raised blade to the cake and remains still during the line.
Dialogue, 0.3-1.8 seconds:
A says: "これで、おわりだ！"
Voice timbre reference Audio 1. Lip sync. Speak once in Japanese, confidently and clearly. Finish before 2.0 seconds, then close A's mouth.
Instrumental low strings and percussion build beneath a magical hum. Hard cut.
Shot 2 (4.0-10.0 seconds): THE CAKE CUT
One uninterrupted three-quarter medium shot. Show A's grip and sword, B's hands, the entire cake and tabletop together. Steady camera.
4.0-6.5 seconds: A holds the sword overhead. B pushes the cutting board's edge, sliding board and cake together to the center of the table, directly beneath the blade. The table remains fixed.
6.5-7.2 seconds: B releases the board and withdraws both hands fully to her side. Briefly hold with her hands clearly outside the blade's path.
7.2-8.5 seconds: Only after B's hands are clear, A performs exactly one powerful downward stroke. Feet planted, two-handed grip consistent. A short, brilliant teal arc follows the blade.
Show the blade physically contacting the cake before it separates.
The blade cuts through the cake's center and stops at the board. Exactly two equal semicircular halves remain upright on the board, with a narrow gap and visible fresh sponge cross-sections. The board and table stay intact.
8.5-10.0 seconds: A lifts the blade slightly to reveal the clean cut, then holds still. The teal light fades. B calmly admires the neatly divided cake.
Audio: one sharp sword whoosh, then a soft cake-slicing sound. Stop the dramatic music and magical hum abruptly when the two halves are revealed. No dialogue. Hard cut.
Shot 3 (10.0-12.0 seconds): JUST DESSERT
Medium two-shot, A left and B right. Keep both cake halves visible.
B casually picks up the fork from the saucer. A small ceramic clink. She looks at the cake with her usual calm expression and closed lips.
A looks from the cake to B's fork. Her confidence disappears. Her sword points downward beside her body.
No dialogue. Fork clink and faint rooftop ambience only. Hard cut.
Shot 4 (12.0-15.0 seconds): THE FAMILIAR FACE
Close-up of A, preserving Image 1's facial design.
Eyes widen, pupils become small, eyebrows rise. Lips part slightly once, then remain still. Her familiar stunned, uncomprehending expression.
A short, controlled push-in. Hold her face for the final full second, with only tiny shoulder trembling.
No dialogue or vocal reaction. End directly on A.
AUDIO CONTROL
Exactly one spoken sentence: A's line in Shot 1. After 2.0 seconds, both characters produce zero vocal sound.
Keep the specified instrumental music, sword effects, fork clink and ambient wind.
No improvised speech, narration, inner monologue, gasps, grunts, screams, laughter, vocal breaths, choir or vocal samples. Never replay Audio 1's recorded words.
NEGATIVE PROMPT
Extra characters, duplicate bodies, mixed identities, costume changes, realistic character faces, 3D game-CG, chibi, extra limbs or fingers, fused hands, unstable props, repeated swings, detached energy beams, crushed or exploding cake, extra cake pieces, broken table, building destruction, injury, wedding staging, subtitles, readable text, logos, watermarks, HUD, end card, canned laughter, comedy sting, excessive camera shake.
Create a 30-second cinematic Japanese idol pop music video featuring two adult female idols. Use energetic micro-cuts, rapid visual variation, and frequent transitions synchronized precisely to an uplifting, high-tempo idol pop beat.
Idol A: adult Japanese female idol, long glossy black hair, bright expressive eyes, elegant and cool presence.
Idol B: adult Japanese female idol, long honey-blonde hair with soft peach-pink highlights, warm confident smile, playful charm.
Keep both faces, hairstyles, and identities consistent throughout the video.
[00:00–00:03]
Extreme close-up of Idol A’s eye. Micro-cuts: eye reflection, lips forming a slight smile, hair moving in soft wind, fingers touching a silver earring, white sneaker stepping onto a glowing stage. Pearl-white studio, pastel-blue shadows, soft beauty lighting.
[00:03–00:06]
Fast graphic transition. Flash cuts between glossy cherry-pink ribbons, light-blue translucent hearts, gold star charms, rotating compact mirrors, and a handwritten pop title card. Idol B appears in a white tailored idol outfit with pink accents. Quick close-ups of her smile, hair flip, wink, and a small dance pose.
[00:06–00:10]
Rapid alternating portraits of both idols. Switch every 0.4–0.8 seconds between:
- high-fashion white studio portraits
- pastel-pink checkerboard set
- reflective sky-blue stage
- close-up of sparkling accessories
- feet landing in synchronized choreography
- camera orbit around both idols
- overhead shot of a heart-shaped stage light
Use smooth whip pans, flash frames, match cuts, and rhythmic jump cuts.
[00:10–00:14]
Both idols perform synchronized idol-pop choreography on a bright pastel stage. Micro-cuts between full-body dance, hand gestures, smiling faces, side profiles, spinning skirts, moving spotlights, and reflections on the glossy floor. Include compact but clear choreography rather than long continuous dancing.
[00:14–00:17]
A rapid fashion-change montage. Idol A alternates between a pearl-white jacket look, a sky-blue short dress, and a refined navy-and-gold idol costume. Idol B alternates between a cherry-pink jacket look, a white-and-gold dress, and a soft peach stage costume. Use snap transitions and motion-match cuts. Keep facial identity unchanged.
[00:17–00:20]
Idol B reaches toward the camera with an energetic smile. Cut instantly to Idol A doing the same gesture from a contrasting blue-lit set. Alternate close-ups, wide shots, lens flare, light leaks, floating glitter, and a brief overhead shot. Shallow depth of field, sharp faces, soft cinematic glow.
[00:20–00:23]
Fast pop-art typography sequence. Huge physical glossy 3D letters reading “SHINE ON” move through a pearl-white and cherry-pink space. Letters rotate, slide, and pass beside the idols without covering their faces. Insert micro-cuts of stars, ribbons, hands, smiles, sneakers, and camera-flash bursts.
[00:23–00:26]
Both idols dance in a stylized split-screen composition. Left side: pearl-white and sky blue. Right side: cherry pink and warm gold. The split changes rapidly from vertical to diagonal to four small panels. Show synchronized gestures, close-up smiles, playful glances, and one clean matching final dance move.
[00:26–00:28]
High-energy climax. Enormous glossy 3D typography reading “MAKE A WISH” rushes toward the lens. Quick flashes of both idols in all three costume looks, glittering stars, heart lights, satin ribbons, rotating mirrors, and bright stage spotlights.
[00:28–00:30]
Final centered close-up: both idols stand together in their pearl-white and gold main costumes. They look directly into the lens with radiant confident smiles. Huge bold typography reading “SPARKLE” appears behind and partly around them. End with prism light leaks, gentle chromatic glow, fine film grain, and a sharp clean white flash.
Visual style: premium Japanese idol pop music video, sophisticated kawaii fashion editorial, cheerful and aspirational mood, pearl white, cherry pink, sky blue, warm gold, and small navy accents. Ultra-detailed adult faces, realistic skin texture, glossy cinematic lighting, shallow depth of field, clean synchronized choreography, polished pop-art typography, fluid camera movement, luxury commercial quality.
photorealistic, cinematic, highly polished, no watermark, no subtitles, no visible text errors.
Original Wan 3.0 Text-to-Video Prompt (10-second duration)Duration: 10 seconds
Aspect ratio: 16:9
Style: Hyper-realistic cinematic sci-fi, photorealistic space photography mixed with ethereal cosmic fantasy, ultra-detailed 1080p, smooth continuous motion, no cutsStep-by-step structured prompt (copy-paste ready):Global scene setup
In the infinite void of deep space, a vast, luminous nebula of swirling violet, emerald, and molten gold gas clouds drifts like living silk. Countless pinprick stars glitter against the blackness. Delicate ribbons of iridescent cosmic dust catch dual light from a distant binary star system—one cool sapphire, one warm amber. Tiny ice asteroids tumble slowly in the mid-distance, their crystalline surfaces refracting rainbows. Soft volumetric light rays pierce the nebula, creating god-rays that feel almost tangible.Subject
An original alien spacecraft of pure biomechanical design floats at the center: the “Aether-Serpent.” It is elongated and asymmetrical, resembling a living deep-sea predator fused with crystalline architecture. Its hull is covered in overlapping, iridescent scale-plates that shift between deep indigo, opalescent silver, and translucent violet. Organic, vein-like conduits pulse with internal bioluminescent cyan energy. Three pairs of graceful, wing-like solar sails made of semi-transparent membrane stretch outward, currently torn and flickering. At the rear, thruster-gills resemble living orifices that once exhaled plasma. No human or Earth-like technology is present—everything feels ancient, extraterrestrial, and slightly alive.Action & fire (core motion)
The Aether-Serpent is critically damaged and on fire. From a jagged rupture along its midsection, ethereal blue-white plasma flames and molten crystalline shards erupt in zero-gravity. The “fire” does not behave like Earth fire: it blooms in slow, elegant spheres and ribbons of glowing plasma, sparks of liquid light, and drifting embers of pure energy that float outward like fireflies. Glowing cracks race across the scale-plates, releasing soft pulses of cyan light. Small fragments of hull and membrane break away and drift in graceful arcs. The entire craft slowly tumbles end-over-end, rotating once fully during the 10 seconds, while the plasma fire intensifies then softens in rhythmic language
Single continuous 10-second take, no cuts.
Start: medium-wide shot, slightly low angle, 35 mm lens feel.
Camera performs a slow, elegant orbital tracking move clockwise around the craft (approximately 90 degrees of orbit), simultaneously pushing in gently from wide to medium-close.
Shallow depth of field keeps the spacecraft razor-sharp while the nebula and distant stars remain softly dreamy. Subtle lens flare from the binary stars drifts across frame as the camera moves.Timeline (exact 10-second breakdown) [0–3 s]: Wide establishing shot. The damaged Aether-Serpent drifts into frame, plasma fire just beginning to bloom from the rupture. Slow orbital camera starts. Ambient cosmic stillness.
[3–7 s]: Camera continues orbiting and pushes closer. The craft completes half its tumble. Plasma flames expand into beautiful, zero-g spheres and ribbons; molten crystal shards and glowing scale fragments float away in elegant trajectories. Bioluminescent veins pulse brighter.
[7–10 s]: Medium-close as the craft finishes its rotation. Fire reaches a soft peak then begins to calm into lingering embers. Final gentle push-in reveals the intricate scale texture and drifting debris against the glowing nebula. Fade on the beauty of destruction.
Lighting & atmosphere
Dramatic dual-source lighting from the binary stars + self-illumination from the plasma and bioluminescence. High contrast between the cold void and warm-cool plasma glow. Soft volumetric god-rays through the nebula. Subtle film grain and anamorphic lens characteristics
Deep, resonant low-frequency space hum mixed with crystalline crackling of plasma fire
GLASGOW — AFTER THE RAIN | 15-Second Cinematic City Tour | WAN 3.0 Video
Create a hyper-realistic 15-second cinematic fashion and city-tour film set in Glasgow, Scotland, progressing from overcast late afternoon into early evening. Use the supplied image as the opening frame and the definitive reference for both characters, their faces, hairstyles, body proportions, outfits, and relationship.
The couple remains together and clearly visible throughout all five scenes. Preserve their natural chemistry, recognizable identities, and consistent clothing. Glasgow must feel authentic, populated, and geographically recognizable.
Scene 1 | 0–3s — Buchanan Street
Begin with the supplied image. Both characters briefly hold direct eye contact with the camera for less than one second, then naturally walk forward together. His arm rests around her waist. The camera tracks backward at walking speed. Pedestrians pass, storefronts glow, and rain-darkened pavement reflects the surrounding city. Natural movement, confident expressions.
Scene 2 | 3–6s — Royal Exchange Square
Hard cut to the couple walking past the Gallery of Modern Art, with the Duke of Wellington statue and its distinctive traffic cone visible behind them. Smooth lateral camera movement. She notices the cone, smiles, and playfully nudges his arm. He follows her glance and laughs. Genuine spontaneous interaction, with pedestrians crossing the square.
Scene 3 | 6–9s — Ashton Lane
Hard cut to the couple entering Ashton Lane beneath its illuminated overhead lights. The camera moves close alongside them, passing briefly near their shoulders. A woman walking in the opposite direction smiles and exchanges a quick friendly greeting with the female protagonist, who smiles back. Cobblestones, pub windows, and small groups socializing create authentic Glasgow evening activity.
Scene 4 | 9–12s — Finnieston
Hard cut to lively Argyle Street in Finnieston. The couple walks past a busy independent restaurant. He briefly holds the entrance door for a departing customer, who acknowledges him with a nod. She takes his hand and leads him onward. Dynamic close-range tracking captures their hands, expressions, and movement before settling into a medium two-shot. Real restaurant activity continues naturally behind them.
Scene 5 | 12–15s — Kelvingrove
Hard cut to a beautiful early-evening street-level view of Kelvingrove Art Gallery and Museum. The couple walks together, with its recognizable red sandstone architecture in the background. He puts his arm around her shoulders. She looks up at him, smiles, then turns toward the camera. A fluid medium-close tracking shot holds their faces for a strong, intimate final portrait. End naturally while they are still moving.
Cinematography & Performance
Premium editorial fashion-film photography. Five distinct scenes connected by clean hard cuts. Alternate backward tracking, lateral movement, close-range passes, and intimate medium portraits. Natural 35mm and 50mm lens aesthetics, believable handheld motion, realistic depth of field.
Keep both protagonists visually prominent in every shot. The city provides context and atmosphere without overtaking the characters. Expressions are understated, spontaneous, and human. Natural walking speed, realistic gestures, physical contact, and genuine reactions.
Lighting evolves subtly from soft overcast daylight to early-evening illumination, with practical storefront and restaurant lights. Wet streets and autumn atmosphere. No nighttime darkness or exaggerated golden-hour lighting.
Continuity & Realism
Preserve exact facial identities, hair, proportions, and wardrobe across every scene. Reconstruct each Glasgow location accurately from the required camera position. Authentic Scottish architecture, recognizable landmarks, credible street layouts, realistic signage, and believable pedestrian behavior.
No morphing, facial drift, duplicated characters, disappearing objects, teleportation, unnatural hand movements, floating bodies, fake skin, plastic textures, or excessive smoothing. No slow motion, zooms, empty establishing shots, artificial lighting effects, subtitles, titles, or generated music.
Natural ambient city sounds and brief conversational greetings only.
15-second cinematic lifestyle and fashion film. Six scenes, clean hard cuts, photorealistic cinematography, natural human performances. WAN 3.0 video.
Character & Visual Continuity
Begin with the exact supplied reference image as the opening frame. Preserve the man's facial identity, athletic physique, tousled brown hair, faded mauve graphic T-shirt, oversized khaki cargo shorts, white crew socks, grey sneakers, and arm tattoo throughout.
Keep his appearance consistent across all six scenes. Natural skin texture, realistic facial expressions, accurate anatomy, and physically believable movement. No artificial skin smoothing, identity drift, morphing, or wardrobe changes.
Scene 1 — Ocean Drive | 0:00–0:02.5
Begin with direct eye contact matching the reference image for less than one second. He pushes away from the sticker-covered railing and walks toward the camera with relaxed confidence. The camera tracks backward at close range, then drifts laterally to reveal Ocean Drive's palms, colorful Art Deco buildings, pedestrians, and passing traffic. Strong morning sunlight creates crisp palm shadows across the pink sidewalk.
Scene 2 — Art Deco District | 0:02.5–0:05
Hard cut to a moving medium shot as he walks past the pastel Art Deco façades of Ocean Drive. A friend approaching from the opposite direction recognizes him. They exchange a quick smile, casual greeting, and natural passing hand slap without stopping. The camera sweeps alongside him, catching the interaction before following his forward movement. Café patrons and pedestrians remain active in the background.
Scene 3 — Sidewalk Café | 0:05–0:07.5
Hard cut to a lively outdoor café along Ocean Drive. He approaches the counter and accepts a freshly prepared iced coffee directly from a smiling barista. Their hands make believable contact with the cup during the exchange. He gives a brief appreciative nod and continues walking. A close lateral camera pass captures his face, hands, and shirt texture, with café activity visible behind him.
Scene 4 — Lummus Park | 0:07.5–0:10
Hard cut to the palm-lined pedestrian paths of Lummus Park. He walks toward the beach while a cyclist passes behind him and two runners move along the path. The camera tracks beside him at waist height, then arcs forward into a medium-close composition. He glances toward the beach with a subtle smile. Sunlight filters through palms, casting moving shadows across the pathway.
Scene 5 — South Beach Promenade | 0:10–0:12.5
Hard cut to the beachside promenade near a colorful Miami Beach lifeguard tower. He walks past two friends carrying surfboards. One casually greets him, and he responds with a friendly nod and quick spoken "What's up?" The camera makes a fast, controlled close-range pass across his shoulder, then settles into a moving three-quarter portrait. Ocean, sand, beachgoers, and palm trees remain clearly recognizable.
Scene 6 — Miami Beach | 0:12.5–0:15
Hard cut to the beachfront with the turquoise Atlantic Ocean behind him. He walks a few steps toward the camera, stops naturally, shifts his weight, and looks directly into the lens with a confident, understated smile. The camera moves gently into a medium-close portrait, retaining the recognizable beach environment. Finish on a strong editorial fashion frame in bright morning sunshine.
Cinematography & Performance
Dynamic, cinematic handheld and stabilized tracking shots. Alternate forward tracking, lateral movement, close-range passes, and intimate portrait framing. Every scene features the protagonist prominently. Maintain purposeful walking motion between cuts and believable human interactions.
Use natural Miami morning colors, authentic Art Deco architecture, accurate South Beach geography, realistic street activity, and consistent directional sunlight. Sharp, believable palm-tree shadows. Natural motion blur and realistic fabric movement.
Clean hard cuts only. No dissolves, transitions, artificial slow motion, zooms, empty establishing shots, or sudden camera jumps. No duplicated people, unnatural hand movements, floating objects, or disappearing props.
Audio
Natural environmental sound only: light traffic, footsteps, café conversations, ocean breeze, distant beach activity, and brief human greetings. No music, soundtrack, singing, or background score.
Create a 15-second, 16:9, 30fps premium compact-camera concept ad as one continuous visual narrative around a single unbranded portable camera. Product design: horizontally elongated rounded rectangle, about 2.1:1; large circular lens on the left, about 68% of body height; narrow pure-black display area on the right; silver sandblasted matte-aluminum frame; black micro-textured body; flush circular top shutter; tiny dark-red front status dot; precise continuous edge seam. Lens: six dark-gray aperture blades, brushed-black ring, and layered dark optical glass with subtle blue-gray/cool-white reflections only—no rainbow flare. Show no brand, logo, text, numbers, or settings. Keep this exact design: never turn it into a phone, DSLR, or action camera; add no lenses, buttons, screens, grips, or flash.
Style: premium industrial-design advertising and precision electronics photography; restrained framing, generous negative space, and fast, exact camera moves. Light with a large rectangular soft source upper-left, a thin rim upper-right/rear, and a controlled vertical soft reflection on the lens. Backgrounds shift between black, cool gray, and warm white. Four consistent minimalist art stills: an ivory feather floating on deep graphite gray; black volcanic rock on warm-white seamless; curved silver sheet as minimal sculpture; circular ripples on cool-gray water. Palette: black, white, silver-gray, with only a hint of blue-gray in water. No people, landscapes, flowers, pets, or cluttered desks.
Timeline:
0–1.7s — Extreme macro inside the lens. Six aperture blades rotate outward together; the opening grows from a tiny cool-white point to about 30% of frame height. Pull back rapidly along the optical axis, revealing credible internal lens depth and mechanics.
1.7–3.5s — Continue pulling back while arcing about 25° right. A thin edge light sweeps across and reveals the full body at about 55% frame width.
3.5–5.3s — Rush into the lens center and transition to a warm-white studio. The viewfinder contains only the floating ivory feather between deep gray and warm white, plus four corner marks, a hollow central focus box, thin horizontal line, and small red dot—no text/settings. The box contracts to lock focus as the camera arcs about 12° around the feather for parallax.
5.3–7s — Fire the shutter without a white flash. Aperture closes about 40% inward, briefly dimming exposure. The viewfinder becomes a refined image card with a thin warm-white border; three matching cards fan out behind it, showing rock, metal, and water. All cards have identical size and spacing and float in warm white.
7–9.2s — Cards form a horizontal sequence. Travel quickly left-to-right at low height; nearby edges gain directional blur while distant cards create parallax. Pass feather, rock, and metal, then brake before the water card with a 5% overshoot and settle.
9.2–11.4s — Water card enlarges about 8% and ripples. Accelerate into it until water fills the screen; skim about 3 cm above the cool-gray surface as circular ripples expand and the central highlight gradually assumes the lens shape.
11.4–13.2s — Approach the dark circular reflection at the center. As it fills frame, shape-match cut to the product’s front lens glass. Pull out and arc about 30° up-left to reveal the full camera while the background becomes seamless warm white.
13.2–15s — Product floats at a three-quarter front-left angle. Arc about 20° upward and push in about 12%, decelerating with a slight overshoot and settle. A cool-white highlight crosses the frame and lens ring; the red dot lights once. Hold the final 0.6s perfectly still, product about 48% frame width in generous warm-white negative space; no title or fade to black.
Use shape/position matches among lens, cards, ripples, and product lens. No harsh cuts, flashes, liquid morphs, or impossible deformation. Sound: restrained, precise modern experimental electronics—blade friction, mechanical lock, low air-suck on pullback, focus ping, clean shutter click, paper-like card unfolding, stereo card-pass whoosh, low water immersion, ripple resonance, and subdued final product thump. No narration, vocals, orchestral music, or heavy EDM drums.
Avoid cheap plastic or toy-camera styling, rough chamfers, mirror-like metal, rainbow flare, blue-purple tech glow, neon, and cyberpunk. Avoid thick Polaroid borders, random card rotation, intersections, duplicated images, distortion, or disappearing elements.
Create a 30-second cinematic near-future social commentary film with one coherent visual language: sleek photorealistic megacity, cool cyan and steel-blue light, restrained camera movement, subtle electronic ambience, realistic human behavior, no logos, no readable brand names, no on-screen captions.
Structure the film as six connected beats with clear visual transitions and escalating unease: 0-5s, autonomous delivery drones and small sidewalk bots bring steaming meals to apartment doors while residents receive them through smart lockers; 5-10s, people summon a silent autonomous taxi from their phones, the vehicle arrives with an empty driver’s seat and smoothly carries them through neon traffic; 10-15s, anxious investors sit before transparent AI financial interfaces asking for advice, algorithmic recommendations replace human brokers, no brokers or advisors are present; 15-20s, a wide public square and office district reveal people with no jobs, automated systems operating everywhere while unemployed adults wait quietly, subdued faces and empty workplaces; 20-25s, an enormous clean factory shows robotic arms producing identical consumer products, packaging them and lining them up in immaculate rows, no human workers visible; 25-30s, pull back to a vast silent retail warehouse and storefronts filled with untouched products, no customers buying, a few people stare through glass at the abundance, the final image is still and emotionally cold. End with a slow wide shot emphasizing that automation has optimized production and convenience but removed work, human mediation, and demand.
Avoid gore, violence, comedy, or exaggerated sci-fi spectacle; make the critique feel plausible, quiet, and unsettling.
The attached video guides dynamic framing, pursuit, counterattacks and impact rhythm.
Image 1 defines the orange-haired origami heroine.
Image 2 defines the black-haired umbrella heroine.
Preserve their exact character designs, stylized proportions, costumes, colors and signature props across every shot. Replace the water setting with a dry clock tower at sunset.
STORY
At the final sunset bell, a magical clock core will seal the tower. The orange-haired heroine must retrieve it before the last chime. The umbrella heroine guards it, convinced that removing it is a mistake. Their opposing goals drive every action. Open in motion, reveal the objective visually, and finish with a decisive reversal.
ABILITIES
Origami heroine: fast footwork, acrobatic changes of direction and ONE red-and-cream bird-shaped returning blade matching image 1. Its curved flight is deliberate. Track this single weapon continuously; it cannot appear in her hand while flying.
Umbrella heroine: precise counters, opening the reference canopy to block, closing it for staff-like exchanges, drawing short-lived ink barriers with its pen-shaped tip. Maintain the umbrella's exact handle, tip and canopy.
LOCATION
A vast old clock tower, amber light through its monumental glass dial, rotating brass gears, broad stone platforms and a massive pendulum. A small golden clock core floats above a pedestal behind the umbrella heroine. Establish the geography clearly and keep it consistent.
0–4 SECONDS
A bell strikes. Close-up of the golden core and the mechanism beginning to close around it. Immediately track the origami blade racing toward the pedestal. The umbrella heroine steps across its path and snaps her canopy open, deflecting it behind the swinging pendulum. Orange hair is already sprinting after her throw.
4–10 SECONDS
Low lateral tracking as orange hair crosses a gear walkway, ribbons snapping. The umbrella heroine closes her canopy and intercepts.
Three readable exchanges: shaft sweep and leap; descending kick and shaft block; quick forward jab and twisting sidestep. Show anticipation, contact and recovery. Orange hair remains unarmed while her blade is away. Every move brings her closer to the core or forces her back.
10–15 SECONDS
The pendulum sweeps between them. Orange hair uses its passing silhouette to conceal a sudden change of direction, vaulting toward the pedestal.
Her opponent anticipated this: the pen tip draws an ink line across the stone. A curved ink barrier rises directly ahead. Orange hair lands, brakes hard and skids; the path to the core closes.
15–20 SECONDS
Another bell. The core's enclosure narrows.
Close-up: orange hair looks trapped, then smiles and flicks two fingers.
Reveal the opening throw paying off: the SAME paper blade has curved around the pendulum and now approaches the pedestal from the opposite side.
The umbrella heroine notices its reflection in the brass mechanism. Her eyes sharpen. She must turn away from orange hair to intercept it.
20–25 SECONDS
She pivots, opens her umbrella and deflects the returning blade. The ink barrier weakens as her attention shifts.
Orange hair bursts through the fading opening, springs from a low gear housing and visibly catches the blade by its grip.
She immediately redirects her momentum toward the core. The umbrella heroine slides across her route. One sharp close-range clash of blade and umbrella shaft, a brief graphic impact frame, then a wide shot restoring their positions.
25–30 SECONDS
Orange hair releases the blade again, making her opponent raise the canopy. It is a feint: the paper bird folds its wings and slips around the canopy toward the closing mechanism.
It wedges between two brass teeth for one crucial instant.
Orange hair reaches through and pulls the core free just before the final bell.
The mechanism stops. Sudden silence.
They land on opposite sides of the pedestal. Orange hair holds the core; the returning paper bird settles into her other hand. The umbrella heroine lowers her canopy, surprised but composed. End on their unresolved stare in the sunset.
ANIMATION AND SOUND
Classic cel-anime aesthetics: clean expressive lines, flat colors, strong silhouettes, bold perspective and selective smear drawings. Fast 1–3-second shots with clear action continuity. Use only very brief holds at key reversals. Alternate close combat framing with wide spatial views. No random teleportation, prolonged slow motion or generic beam exchanges.
Escalating orchestral-rock music, ticking gears, bell strikes, snapping canopy, blade whistles and sharp magical impacts. Drop the music at the final stopped mechanism. No dialogue, captions, logos, gore or costume changes.
Style: Pixar-style 3D animation, warm rainy-day palette, soft cinematic lighting.
A rainy city street at dusk, neon reflections shimmering on wet pavement. A small child walks holding a bright red umbrella. A sleek black crow swoops down, camera following in a quick tracking shot, snatches the umbrella handle in its beak, and flies off with it — child gasps, reaching up in surprise. Cut to the crow flying low over rooftops, umbrella flapping behind it like a cape, rain streaking past camera. It dives under a rusty bridge where a small shivering stray dog huddles against the cold. The crow drops the umbrella, angling it with its beak to shield the dog from the rain. Close-up on the dog’s grateful eyes, tail thumping. Final wide shot: crow perched on the umbrella’s edge, both animals dry together as rain falls softly around them. Sound: gentle rain patter, distant city hum, soft orchestral swell on the final shot, no dialogue.
Create a stunning cinematic anime sequence set in a magical Japanese-inspired coastal town at sunset.
A young anime girl with long flowing dark hair, expressive eyes, and a beautiful white dress stands on a quiet seaside cliff overlooking the ocean. Her hair and dress move naturally in the warm evening breeze.
The video begins with a wide establishing shot of the glowing coastline, colorful clouds, distant mountains, and sparkling ocean. The camera slowly pushes toward the girl as she looks toward the horizon.
She gently closes her eyes, then opens them and smiles softly. Small glowing particles begin floating around her, carried by the wind. The camera smoothly circles around her while her hair and dress flow dynamically.
She starts running along the cliffside path toward the sunset. The camera follows her from behind, then smoothly moves beside her, capturing her joyful expression and flowing hair.
As she reaches the edge of the coastal path, she stops and looks toward the enormous glowing sunset. Hundreds of tiny luminous particles rise into the sky around her. The camera slowly pulls backward and upward, revealing the beautiful ocean, coastline, mountains, and glowing sky.
High-quality cinematic Japanese anime aesthetic, detailed hand-drawn anime characters, expressive eyes, beautiful facial animation, fluid character movement, dynamic hair physics, flowing clothing, detailed backgrounds, atmospheric perspective, dramatic sunset lighting, vibrant colors, soft cinematic glow, subtle depth of field, smooth camera movement, consistent character design throughout, polished feature-film anime quality.
No cuts, no flickering, no character morphing, no distorted anatomy, no extra limbs, no deformed hands, no inconsistent face, no text, no subtitles, no watermark.
{
"archetype": "Competitive cooking show finale",
"duration": "30s",
"prompt": {
"concept": {
"title": "The Judge Who Never Speaks",
"description": "A legendary, feared food critic known for never speaking during judging takes one bite of a nervous contestant's dish. His face goes through a slow, dramatic silent transformation while the entire kitchen holds its breath.",
"rhythm_structure": {
"0-8s": "Tense kitchen setup, contestant plating nervously",
"8-13s": "Dish is presented, judge picks up fork in total silence",
"13-15s": "The bite — extreme close-up, no reaction yet",
"15-25s": "Slow silent facial transformation, crew and contestant frozen watching",
"25-30s": "Single word verdict, contestant collapses in relief/joy"
}
},
"character_identity_lock": {
"maintain_same_character": true,
"identity": {
"age_gender": "stern man in his 60s",
"hair": "silver, perfectly groomed",
"face": "deep set eyes, unreadable resting expression, weathered but composed",
"clothing": "immaculate black suit, no tie, single cufflink glint",
"accessories": "antique pocket watch he never uses, always in pocket"
},
"strict_consistency": ["Same stern face throughout", "Never speaks until the final word", "Same suit and posture"],
"color_rule": "Cool steel-blue kitchen lighting throughout; warm golden light isolates his face during the reaction",
"core_concept": "Maximum tension built entirely through silence and micro-expression, not dialogue",
"physics_rules": "Fully realistic, no exaggeration — the drama is in restraint, not spectacle"
},
"camera_direction": {
"shot_type": "Slow deliberate cinematography, extreme close-ups, minimal cuts",
"forbidden": ["No fast cuts during the reaction", "No dialogue until the final word", "No exaggerated cartoonish faces"],
"camera_journey": "Wide steady shot of tense kitchen, slow push into contestant's shaking hands plating, extreme close-up on the fork lifting, held close-up on judge's face for the full reaction, cut wide for the final word and contestant's collapse."
},
"visual_style": {
"overall": "High-end prestige cooking show cinematography, moody and cinematic rather than bright reality TV",
"inspired_by": ["The Bear tension", "MasterChef finale drama", "Ratatouille's critic scene gravity"],
"color_palette": ["Cool steel blue kitchen", "Warm isolated golden spotlight on judge", "Deep black suit contrast"],
"materials_and_texture": "Realistic steam, plate ceramics, fabric sheen, sweat on contestant's brow"
},
"motion_language": {
"follows_music": true,
"tension_strings": "Single sustained note building through the silent reaction",
"silence_beat": "Near-total ambient silence for 8+ seconds during the face transformation",
"release_hit": "Single soft piano note on the verdict word"
},
"storyboard": {
"total_beats": 6,
"beats": [
{"id": "01", "time": "0-8s", "description": "Wide shot of a tense professional kitchen. Contestant's hands shake slightly while plating a delicate dish. Crew watches silently."},
{"id": "02", "time": "8-13s", "description": "Judge sits at the tasting table, expressionless. Dish is placed before him. He picks up the fork with slow deliberate precision."},
{"id": "03", "time": "13-15s",
"description": "Extreme close-up as he takes the bite. Chews once, twice. Absolutely no readable expression yet."},
{"id": "04", "time": "15-25s", "description": "Held close-up on his face as it slowly shifts — eyebrow twitch, eyes widen slightly, jaw tightens, then a barely perceptible softening. Contestant and crew frozen, not breathing."},
{"id": "05", "time": "25-28s", "description": "He sets down the fork, looks up, pauses, then says one single quiet word: 'Exceptional.'"},
{"id": "06", "time": "28-30s", "description": "Contestant's knees buckle in relief, crew erupts silently into hugs. Freeze on the judge allowing the faintest smile."}
]
}
}
}
Create a 15-second ultra-cinematic street racing sequence with the adrenaline and intensity of a modern arcade racing game.
0–3 sec — Launch:
Midnight in a neon-lit city. A modified supercar sits at a red light, engine vibrating aggressively. Close-up of glowing headlights, spinning rims, driver gripping the steering wheel. The light turns green — instant launch with tire smoke and dramatic camera shake.
3–7 sec — High-Speed Chase:
Low-angle tracking shot beside the car as it blasts through wet streets at extreme speed. Neon signs reflect across the glossy bodywork. Rapid gear shifts, glowing dashboard, realistic motion blur, sparks from a near-miss with a barrier. Camera swings from rear chase view to a dynamic side shot.
7–11 sec — Impossible Corner:
The car approaches a sharp city corner at full speed, slams the brakes, then performs a perfectly controlled drift. Tires scream, smoke fills the street, headlights cut through the haze. Camera rotates around the vehicle in a fast 360° cinematic move.
11–15 sec — Final Sprint:
The car exits the drift and rockets toward a massive illuminated finish line. A second rival car appears beside it. Both accelerate side-by-side through a tunnel of neon lights. Final shot: extreme low-angle rear view as the winning car disappears into the night.
Visual style: photorealistic AAA racing-game graphics, cinematic night lighting, wet asphalt reflections, detailed car materials, realistic tire physics, volumetric fog, dynamic motion blur, lens flares, aggressive camera movement, high-speed perspective, dramatic depth of field, premium VFX, seamless transitions, intense racing atmosphere.
Audio: roaring engine, turbo spool, gear shifts, tire screech, bass-heavy cinematic soundtrack, Doppler effects and a powerful final acceleration hit.
Format: 9:16 vertical, 15 seconds, ultra-detailed, realistic physics, consistent car design throughout, no text, no logos, no watermarks.
Create a fast-paced 15-second multi-shot smartphone nightlife vlog featuring TWO young adult Japanese women enjoying a wild, energetic night out in Tokyo.
The concept is: “two stylish girls eating and drinking their way through Tokyo at night.”
The tone is loud, playful, chaotic, glamorous, and fun — like spontaneous camera-roll footage from a real girls’ night out.
NOT a commercial.
NOT a polished fashion film.
NOT influencer posing.
Use short 1–2 second clips connected by natural hard cuts.
CHARACTERS
Exactly TWO young adult Japanese women appear throughout the entire video.
They are close female friends in their early 20s, both very stylish, confident, attractive, and full of energy.
WOMAN A:
A beautiful young adult Japanese woman with shoulder-length light-brown hair, soft wispy bangs, fair skin, and natural but slightly glamorous makeup. Slim feminine figure.
WOMAN B:
A beautiful young adult Japanese woman with long dark-brown hair, fair skin, and natural but slightly glamorous makeup. Slim feminine figure.
They must remain clearly distinguishable and consistent throughout every shot.
WARDROBE
Both women wear nightlife outfits built around the same styling theme:
* a metallic silver halter top
* a very short black skirt
* long black boots
The outfits should feel bold, sexy, trendy, and nightlife-appropriate, but still believable for a fun Tokyo girls’ night out.
They may have slightly different skirt cuts or accessories, but both should clearly match this same overall look.
No random outfit changes.
CAMERA STYLE
Authentic smartphone footage.
The women film themselves and each other.
Alternate naturally between:
* selfie shots
* one friend filming the other
* both women in frame together
* quick handheld street shots
* spontaneous close-ups of food and drinks
Natural handheld shake, imperfect framing, slight motion blur, occasional autofocus adjustments, accidental tilts, brief partial frame-outs, and late camera reactions are welcome.
The video should feel spontaneous, noisy, messy, and real — like favorite clips pulled from a real phone camera roll.
FORMAT
15-second multi-shot montage.
Use about 8 short shots.
The energy should build quickly and stay high from beginning to end.
SHOTS
0–2s — START OF THE NIGHT
Tokyo at night, bright neon lights, crowded street, likely Shibuya or Shinjuku.
Woman A holds the smartphone in selfie mode while walking with Woman B beside her.
Both are already dressed in their silver halter tops, short black skirts, and long black boots.
They are laughing and full of energy.
Woman A excitedly says in Japanese:
「今日食べまくるよ！」
Hard cut.
2–3.5s — FIRST DRINK
Outside a casual standing bar or small izakaya.
Woman B holds two drinks toward the camera.
Tokyo nightlife glows behind them.
They quickly clink glasses.
Woman B says:
「乾杯！」
They take a sip and laugh.
Hard cut.
3.5–5s — STREET FOOD
Close handheld shot of one of them eating hot skewers or grilled street food.
The other woman leans into frame, trying to steal a bite.
The first woman pulls it away and laughs:
「ちょっと、それ私の！」
Hard cut.
5–6.5s — WALKING THROUGH NEON TOKYO
Both women walk through a lively neon-lit street packed with signs, people, and nightlife energy.
One films from slightly behind while the other turns back dramatically, smiling.
The silver tops catch the city lights.
They are loud, playful, and moving quickly.
Hard cut.
6.5–8s — CONVENIENCE STORE STOP
Inside a brightly lit Tokyo convenience store.
One woman holds snacks, the other holds canned drinks.
They both laugh at the camera like they are making a chaotic late-night decision.
Woman A says:
「これも買う！」
Hard cut.
8–10s — EATING ON THE STREET
Outside again.
Both women stand on the street eating while holding their drinks.
One has full cheeks because she is mid-bite.
The other laughs at her and points the camera closer.
They look a little more messy and hyper now.
Hard cut.
10–12s — RAMEN
Inside or just outside a late-night ramen shop.
A quick close shot of steaming ramen.
Then a shot of both women slurping noodles, laughing, and reacting to the heat.
One says:
「うまっ！」
The other laughs with food still in her mouth.
Hard cut.
12–13.5s — CHAOTIC NIGHT ENERGY
Back outside in neon Tokyo.
The women are louder, more excited, and slightly chaotic now.
One films while the other spins once in the street and poses for half a second, then breaks character and starts laughing.
They bump into each other and keep walking.
Hard cut.
13.5–15s — FINAL SELFIE
Close selfie shot with both women tightly in frame.
Tokyo neon lights blur behind them.
They are holding food and drinks, laughing hard, hair slightly messy, makeup still good but less perfect, looking like they have had the best night.
Woman B shouts in Japanese:
「東京の夜最高！」
Both laugh loudly.
One reaches toward the camera and covers the lens.
End.
PERFORMANCE
The women should feel like real close friends on a fun night out.
Natural laughter, teasing, talking over each other, fast reactions, casual body language, and spontaneous energy.
They should look like they are genuinely having fun eating, drinking, and wandering through Tokyo at night.
All spoken dialogue must be natural Japanese.
Keep dialogue short, casual, and lively.
VISUAL STYLE
Ultra-realistic smartphone nightlife footage.
Tokyo at night should feel dense, bright, and alive:
neon signs, izakaya lights, convenience store lighting, street reflections, crowds, narrow streets, and late-night food spots.
The silver tops should reflect city lights beautifully.
Food should look delicious, hot, and real.
The overall vibe should feel glamorous but messy in a fun way — a stylish girls’ night with lots of food, drinks, and laughter.
LOCKS
Exactly TWO women throughout.
Both are young adult Japanese women in their early 20s.
Both wear:
* metallic silver halter tops
* very short black skirts
* long black boots
No random outfit changes.
Maintain consistent faces, hairstyles, and body proportions.
No third main character.
No polished commercial look.
No perfect stabilization.
No perfect influencer posing.
No artificial transitions.
No morphing.
No CGI appearance.
No text overlays.
No subtitles.
No logos.
The final video should feel like 15 seconds pulled directly from two stylish girls’ chaotic, delicious, neon-soaked Tokyo night out.
STYLE REFERENCE: A high-fidelity, next-gen open-world video game
visual style — rendered like modern AAA game-engine graphics rather
than live-action footage. Semi-realistic character models, real-time
dynamic lighting, physically-based textures, in-game animation
quality, gameplay-capture camera behavior — not film grain, not
handheld imperfections
DURATION: 30 seconds, mostly continuous, minimal cuts
ASPECT: 16:9
SETTING: A city street leading up to a bank branch (generic
architecture, no real bank names/logos), then a car chase and aerial
pursuit
CHARACTERS:
- MAIN CHARACTER: @image1 — exact locked visual identity, reinterpreted
as a game-rendered 3D model, same face/hair/proportions from every
angle. She is an elderly grandmother — voiced with a creaky, raspy,
spirited old-lady voice for all lines. Holding a rifle, later a
bazooka
- GUARD: generic bank security guard, uniformed
- EMPLOYEES: generic bank staff behind the counter
- POLICE: unmarked-style cruisers plus a helicopter with loudspeaker
and spotlight
CAMERA: Third-person, open-world-game-style follow-cam, smooth
programmatic in-game tracking. Behind-the-shoulder during the
robbery, chase-cam once driving. Camera reorients behind each weapon
during aiming, tracking the sightline to the target.
HUD OVERLAY: Small minimap, bottom-left corner, throughout. Subtle
aim reticle appears only while actively aiming a weapon.
BEAT-BY-BEAT:
0:00–0:02 — Grandma driving a car through the city, rifle on the
passenger seat, approaching the bank
0:02–0:03 — Parks directly in front of the entrance, engine idling
0:03–0:04 — Exits holding the rifle, strides toward the door
0:04–0:05 — The guard steps out to intervene; she raises the rifle
and points it at him — he yelps in an exaggerated high-pitched
scream and bolts away in a flailing, cartoonishly panicked sprint,
hat flying off
0:05–0:06 — Kicks the entrance door open with force
0:06–0:07 — Bursts inside, rifle raised, shouting in her creaky
grandma voice: "THIS IS ROBBERY!" — employees and customers flinch
0:07–0:08 — Tosses a duffel bag onto the counter: "Put all the money
in the bag — now, dearies!"
0:08–0:11 — Chaos at the counter: one employee scoops handfuls of
cash into the bag with shaking hands, another knocks over a coin
tray scattering change across the floor, a third fumbles a stack
of bills nearly dropping them, a manager nervously edges toward a
silent alarm button, all under the rifle's stare
0:11–0:12 — Grabs the full bag and sprints back out
0:12–0:13 — Jumps into the car, tires screech as she peels away
0:13–0:14 — Police cruisers spot the car, sirens blaring, giving
chase directly behind
0:14–0:15 — A police helicopter roars in overhead, spotlight
sweeping down
0:15–0:16 — Its loudspeaker blares, distorted: "SURRENDER NOW,
GRANDMA!"
0:16–0:18 — She reaches into the back seat for a bazooka, leans out
the window, hefts it onto her shoulder, aiming up; the reticle
locks on
0:18–0:19 — Fires
0:19–0:20 — Direct hit — the helicopter erupts into a fireball,
spiraling down trailing smoke
0:20–0:22 — Wreckage crashes in the distance, flames and smoke
0:22–0:24 — A pursuing cruiser swerves hard to avoid falling debris
0:24–0:26 — She tosses the spent bazooka aside, refocuses on driving
0:26–0:28 — Car speeds around a corner, pulling ahead of the cruisers
0:28–0:30 — Final wide shot of the car speeding into the city, smoke
still rising behind from the crashed helicopter
CHARACTER DIALOGUE: "THIS IS ROBBERY!" / "Put all the money in the
bag — now, dearies!" (both in a creaky, spirited grandma voice) —
POLICE LOUDSPEAKER: "SURRENDER NOW, GRANDMA!" — no on-screen
subtitles beyond the HUD
AUDIO: Engine idle then roar, hard door kick, shouted commands in an
old woman's voice, employees' frightened murmurs, coins scattering,
cash rustling, engine peel-out and tire screech, sirens, helicopter
rotor thump, the distorted loudspeaker line, bazooka launch whoosh,
explosion boom with debris scatter
NEGATIVE PROMPT: no on-screen text/subtitles, no real bank names or
logos, no real police insignia, no extra HUD beyond minimap and
reticle, no watermark, no gore beyond the stylized explosion, no
design drift on the main character, no first-person view, no harm to
the guard or employees beyond fear/fleeing, no live-action film
grain, no low-poly/cartoon rendering — high fidelity only
Unplanned vertical smartphone recording, shot by a friend moving with the group, imperfect handheld framing, natural night exposure, occasional motion blur, wide phone lens, realistic deep focus.
Characters
The hero is a young woman with long dark brown hair wearing a fitted emerald-green dress with an open back, small black shoulder bag and flat sandals.
Her three friends: one wearing a burgundy dress, a blonde wearing a champagne-colored mini dress, and a brunette wearing a black sleeveless dress — dressed for a night party.
They are laughing, talking loudly and joking together as they walk.
Setting
A glamorous beachfront promenade at night — a wide stone walkway beside the sea, lined with palm trees, beach clubs, outdoor tables and glowing decorative lights.
Warm amber lamps illuminate the promenade while the dark ocean stretches beside the walkway.
A large illuminated beach club pool sits directly beside the promenade, with turquoise underwater lights reflecting across the surface.
Shot list
0.0s–4.0s — Wide tracking shot, eye level, camera moving backward
The camera moves backward in front of the four women as they walk along the beachfront promenade, framed head-to-toe, laughing and chatting in French; palm trees, beach lights and the dark ocean visible behind them, natural handheld bob.
4.0s–6.0s — Medium shot, eye level, backward track + quick pan
The woman in the green dress suddenly runs toward the illuminated pool beside the walkway, raising both arms as her friends shout and laugh; the camera quickly pans to follow her.
6.0s–8.0s — Medium-wide, camera swings right and tilts down
She reaches the pool edge and leaps forward into the water without stopping; the camera swings right and tilts downward to keep her in frame as she disappears below the surface.
8.0s–10.0s — High-angle wide, looking down at the pool, sudden jolt
The camera looks over the pool edge as she crashes into the turquoise water, creating a large splash that spreads across the illuminated surface; the handheld frame shakes from the excitement.
10.0s–13.0s — High-angle medium, looking down, drifting slightly
She pops back above the water, laughing and wiping water from her face, wet hair stuck across her shoulders as she looks directly toward the camera.
13.0s–16.0s — Handheld tilt up and back down, quick reframe
The camera tilts upward toward her three friends standing at the pool edge, laughing and reacting loudly, then quickly tilts back down toward her as she splashes water toward them.
16.0s–20.0s — High-angle medium, following movement
She swims toward the shallow end, grabs the pool ladder and starts climbing out; the camera follows her upward with a loose handheld tilt.
20.0s–23.0s — Low/close medium at pool level, static handheld
She steps onto the stone deck completely soaked, pauses for a second, then bends forward and pushes her wet hair away from her face while water drips from her dress.
23.0s–27.0s — Medium shot, eye level, loose handheld reframe
She walks back toward her friends laughing, flips her wet hair from one side to the other, while the three women surround her and react with loud laughter.
27.0s–30.0s — Medium shot, eye level, camera settles
She turns toward the camera, places one hand on her hip and gives a playful confident smile while her friends laugh behind her; she throws her wet hair backward as the camera holds to the end.
Audio
Live sync sound — loud laughter and excited shouting in French, surprised screams, a large water splash, pool water movement, wet footsteps, distant waves, palm leaves moving and natural beachfront night ambience.
No music.
Style
Raw amateur vertical smartphone footage, realistic night lighting, warm beachfront lamps mixed with turquoise pool illumination, imperfect handheld movement, natural exposure shifts and authentic spontaneous UGC feeling.
0–5s: Wide shot of a sunny seaside fishing pier. A fisherman pulls up his fishing net and discovers a huge crab tangled inside. The crab suddenly escapes and starts running across the wooden pier.
5–10s: The fisherman drops the net and runs after it, trying to catch the crab before it reaches the edge of the pier.
10–15s: Handheld chase shot follows the crab moving sideways at surprising speed, weaving between fishing buckets, ropes, coolers, and tourists.
15–20s: The fisherman tries to block its path with a fishing net, but the crab quickly changes direction and races underneath a wooden bench.
20–25s: The fisherman crawls under the bench after it, gets tangled in his own fishing rope, falls backward, then quickly gets back up as the crab continues toward the water.
25–30s: The fisherman finally corners the crab near the pier edge. He reaches down carefully, but the crab suddenly grabs his fishing hat and runs away wearing it.
Style: ultra-realistic, cinematic action-comedy, natural seaside daylight, realistic ocean and fishing pier, lifelike crab movement, realistic human reactions, dynamic handheld chase camera, subtle camera shake, shallow depth of field, 35mm lens look, natural colors, documentary realism.
Style: classical oil-painting warmth blended with modern commercial studio photoshoot - visible softboxes, C-stands, backdrops, crew edges. Filmic grade, rich browns, natural contrast.
Lighting: studio softbox key per shot, matching each painting's original lighting mood.
BEAT 1 - EXPRESSIONIST FIGURE. [00:00–00:04] MS. Photorealistic human recreation of a famous expressionist figure — gaunt pale bald man in dark clothing, hands pressed to cheeks in dramatic surprise, wavy orange-sky backdrop on a studio stand, softbox visible at frame edge. Slight push-in. He holds the iconic pose, eyes darting to the off-camera photographer. Photographer (off-screen, cheerful): "Can you relax your face?" He replies flatly, hands still on cheeks: "This IS my relaxed face."
BEAT 2 - POST-IMPRESSIONIST PAINTER. [00:04–00:08] MCU. Photorealistic 19th-century painter with a red beard, wearing a blue coat and a white ear-muff over his left ear, swirling blue-green backdrop, studio umbrella visible. Locked-off. An assistant's hand extends a small microphone into frame from the right. Photographer (off-screen): "Say something into the mic?" The painter glances at the mic held near his covered ear, deadpan: "Wrong ear."
BEAT 3 - RENAISSANCE GODDESS. [00:08–00:12] MWS. Photorealistic classical goddess — long flowing red-gold hair, fully dressed in an elegant flowing white toga, standing on a giant scallop shell prop in a studio, seashore backdrop, large white diffuser visible on the left. Locked-off. She sways slightly to keep balance. Photographer (off-screen): "Can you step off the shell?" The goddess, serene but firm: "It came with the outfit."
BEAT 4 - VICTORIAN MOTHER. [00:12–00:16] MS profile. Photorealistic elderly woman from a famous 19th-century portrait — black dress, white lace cap, seated rigid in profile on a wooden chair against a gray wall backdrop, C-stand visible. Slight push-in. She does not turn her head. Photographer (off-screen): "Could you stand for this one?" She replies without moving, dry: "I've been sitting since 1871."
Audio: lighthearted upbeat acoustic music bed across the full 16 seconds, resolving on the final frame; clean off-screen photographer voice — young, cheerful male; each character with a fitting deadpan voice (Figure: thin and strained; Painter: weary Dutch-accented; Goddess: serene and melodic; Mother: dry elderly). All dialogue lip-synced except the photographer.
Preserve: each subject instantly recognizable as their classical inspiration; studio equipment visible in every shot; no on-screen text, no captions, no subtitles. One camera setup per beat, hard cut between beats.
A surreal high-fashion cinematic scene featuring a pale blonde woman in a luxurious oversized white fur coat, beginning with an extreme close-up as she holds a vintage brass saxophone near her face, then staring intensely into the camera and slowly reaching her hand toward the lens, creating dramatic depth and motion blur; the scene transitions to a wide minimalist white architectural interior where she stands and moves forward with both hands extended toward the camera, followed by elegant shots of two versions of the woman seated in sculptural white lounge chairs, maintaining a mysterious editorial mood, soft diffused lighting, clean monochromatic white palette, realistic skin texture, shallow depth of field, smooth cinematic camera movement, subtle film grain, and high-end fashion campaign aesthetics.
A model with long light-brown hair walks and poses in a luxurious studio setting against a soft beige wall and a towering arrangement of white peonies. She wears a dramatic one-shoulder evening gown in muted gray: a heavily embellished bodice and sheer skirt densely covered in silver sequins and crystal beads, cinched by a slim black belt with a floral buckle, and finished with a long flowing sheer cape that trails behind her. Close-ups highlight the intricate beading, her large sculptural earrings, and bracelet, while full shots show her elegant stance, platform heels, and the gown’s movement as she turns to reveal the back. The clip ends with the Ziad Nakad logo over a soft-focus view of the dress from behind.
REFERENCE PRIORITY
@ image1 and @ image2 exclusively control character identities, costume details, proportions and weapons. Never reproduce the original video's characters or display the character sheets themselves.
CHARACTERS AND FIGHTING STYLES
Fighter A, @ image1: deep-purple hair in one long thick braid, green eyes, leaf ornaments, pale-green leaf cape and split leaf skirt, dark fitted costume, asymmetric legwear, dark ankle boots, and TWO short inward-curving green blades with plum-colored fittings. Preserve the exact left/right costume asymmetry from the reference. Her style is elusive, precise and relentless: lateral footwork, compact slashes, outside-line evasions, one blade redirecting while the other counters. Pale-green blade trails and scattered pink petals briefly trace actual movements; they are not ranged projectiles.
Fighter B, @ image2: short ivory-white hair with a small side braid, amber eyes, sleeveless plum wrap tunic, orange waist sash, black fitted leggings, practical purple boots, and TWO oversized purple-and-bronze armored gauntlets with glowing amber cores. Her style is grounded, explosive pressure boxing: compact guard, jabs, hooks, body-level punches, forearm parries and forceful advancing steps. Amber hexagonal impact flashes appear only at contact. Her gauntlets have weight, and committed punches require visible recovery.
ENVIRONMENT AND ART DIRECTION
A rain-soaked abandoned botanical conservatory opens directly into a wisteria courtyard. Tall oxidized bronze window frames, cracked glass, mossy stone paving, hanging violet blossoms and tangled vines. A few surviving amber garden lamps contrast with cool overcast daylight. The ground is firm stone beneath shallow puddles, not deep water. Establish a central cracked circular planter in the courtyard as a spatial landmark.
Use expressive 2D anime linework, controlled cel shading, painted backgrounds and sharply choreographed animation. Purple, sage green and amber unify the characters and setting. Keep faces and costume colors readable. Petals, broken leaves, droplets and small stone fragments make acceleration and impacts visible.
0–4 SECONDS — IMMEDIATE COLLISION
Open on an extreme low tracking shot of B’s boot driving through a puddle inside the conservatory. Cut directly into a right straight punch toward A. A slips outside the fist and redirects the gauntlet sideways with one curved blade, immediately countering toward the opposite armored forearm with her second blade. B catches the counter on her raised gauntlet. Two distinct contacts, sharp sparks and water spray. A medium two-shot clearly establishes both fighters before the camera accelerates.
4–9 SECONDS — PRESSURE AND COUNTERPRESSURE
Track laterally through the narrow aisle. B advances with a fast three-beat combination: jab, opposite hook, then a short body-level punch. A evades the jab, ducks beneath the hook and checks the final punch against the reinforced exterior of the gauntlet, never stopping its full force head-on. A pivots toward B’s flank and answers with two compact alternating slashes. B turns with her, blocking one blade on each armored forearm. Show attack, contact and recoil within readable shots. The braid, sash and leaf skirt lag naturally behind every direction change.
9–14 SECONDS — BREAKOUT INTO THE COURTYARD
A crosses her blades against B’s forearm guard and tries to turn her off balance. B lowers her center of gravity, steps inside A’s reach and drives forward behind both gauntlets. A disengages and springs backward through the conservatory’s already-broken doorway. B follows immediately. Whip-pan through the opening, then cut wide to reveal the wisteria courtyard and circular planter. A lands in a low sideways slide, boots throwing two thin fans of water; B closes the distance before the spray settles.
14–20 SECONDS — CLOSE CALL AND ENVIRONMENTAL IMPACT
B throws a committed downward diagonal punch. A twists clear at the last moment; the gauntlet strikes the edge of a low stone planter, breaking a small section into visible fragments. One very brief impact hold, then full speed resumes. A uses B’s recovery window to spring off the intact planter edge and attack from the side with a descending curved blade. B raises one gauntlet to intercept and immediately fires a short opposite hook. A rotates away from the hook and lands beside her. Follow the movement in a continuous three-quarter tracking shot so the near miss and landing remain clear.
20–26 SECONDS — RELENTLESS INFIGHTING
Move into the fastest exchange, with both fighters staying within arm’s reach. A feints high with one blade and cuts toward the outside of B’s lower gauntlet with the other. B retracts that arm, steps diagonally inward and uses her forearm to knock A’s weapon arm off line. A turns with the deflection, recovering her blade while checking B’s next fist with the second blade. B advances with one heavy hook; A drops beneath it and pivots out beside the planter. Use a fast partial camera orbit with visible background landmarks, alternating medium action shots with very brief determined-face inserts. Each fighter repeatedly attacks, defends and regains balance; neither simply waits.
26–30 SECONDS — CLIMACTIC CONTACT
A immediately turns back with a committed crossing double-blade strike as B drives forward behind a reinforced two-gauntlet guard. The blades meet the armored exteriors of the gauntlets between their bodies. One sharp graphic impact frame: pale-green arcs intersect a compact amber hexagonal flash. The collision sends a low circular wave through the puddles, lifting petals and loose leaves.
Both fighters recoil and slide apart under the force, retaining their weapons. Pull rapidly back and slightly upward to reveal the courtyard, damaged planter and settling spray. End with both already leaning into their next attack, eyes locked, leaving the duel unresolved and energetic.
PACING AND PHYSICAL CONTINUITY
Most shots last approximately 0.6–1.5 seconds, with a few longer tracking shots to make complete combinations legible. No opening pose montage, transformation sequence, dialogue break or prolonged slow motion. Keep impact holds extremely brief. Intensity comes from committed movement, near misses, counters and environmental reactions, not constant camera shake or obscuring flashes.
Keep exactly two fighters. Preserve hair, facial features, costume asymmetry, blade count and gauntlet construction in every shot. Blades remain rigid and correctly gripped; gauntlets stay attached to their corresponding hands. Show credible footing, weight transfer, joint movement and recovery. Maintain consistent screen direction, using visible camera movement or wide shots when changing sides. No teleportation, weapon morphing, duplicate limbs, beam attacks, gore, costume damage, captions or logos. Frame the combat around faces, weapons and movement.
SOUND
Rapid splashing footfalls, blade-on-gauntlet clangs, heavy armored punches, brief exertion breaths, cracking stone and snapping foliage. Urgent percussion supports the action without burying contact sounds. Drop the music for a fraction of a second before the final impact, then hit with a deep percussive collision and falling water.
Style
Hyper-realistic CG cinematography with detailed facial features, realistic skin and hair, cinematic lighting, soft shadows, physically accurate reflections, SSS, global illumination, RTX ray tracing, 4K HDR quality, and realistic cloth and rigid-body physics.
Scene
An ancient theater stage illuminated by flickering candlelight, surrounded by translucent shadow-puppet screens and hundreds of faceless wooden assassins.
The female hero, based on Image 2, wears a dark-red opera costume with copper bells and oversized flowing sleeves. She remains elegant, cold, and completely composed throughout the battle.
Shot 1 — 0–3s
Low-angle FPV fly-through
The camera moves beneath the theater stage as hundreds of wooden puppet assassins close in.
The heroine calmly draws a tiny pair of black iron scissors from her sleeve, holding them delicately between her fingertips.
Shot 2 — 3–5s
Extreme close-up
She gently blows cold air onto the scissors, creating visible white mist.
The tiny scissors rapidly expand into enormous black iron shears.
She says:
“The performance is over. Time to take the final bow.”
Shot 3 — 5–9s
Rapid 180-degree orbit
She launches into an elegant, theatrical combat sequence inspired by shadow puppetry.
Each movement of the giant shears releases razor-sharp black energy blades that slice through the wooden assassins like paper puppets.
Shot 4 — 9–13s
Fast tracking shot
She moves between the wooden pillars, using the environment to propel herself forward.
The giant shears crack the floor and pillars while wood splinters and debris fill the air.
Her flowing sleeves, hair, and copper bells react naturally to every movement.
Shot 5 — 13–15s
Low-angle wide shot · Dutch-angle dolly-in
She lands and rests the giant shears on her shoulder.
Behind her, the motionless assassins stand covered in precise cuts.
With a small flick of her wrist, the entire theater cracks apart and collapses.
Smoke, fire, and black ash fill the background as the image freezes on her cold, confident expression and fades to black.
Audio
Diegetic and action sounds only: wooden boards, copper bells, cold wind, breath, metallic transformation, weapon impacts, energy whooshes, splintering wood, and collapsing debris.
No background music. No subtitles.
**FORMAT:** 30 seconds, 16:9, high-end real-time video game graphics, GTA VI gameplay style.
### ABSOLUTE RULE: NO HUD OR UI
This is third-person gameplay footage with ALL interface elements disabled. Absolutely NO HUD, minimap, radar, icons, menus, text, subtitles, interaction prompts, button indicators, overlays, or customization interfaces. NO UI at the beginning, during hairstyle changes, or at the end. The entire video must remain completely free of visible interface elements. Hairstyle changes are accompanied ONLY by audio confirmation sounds.
### REFERENCES
* @Image-1: Exact opening frame. Start the video with this image, preserving its composition and camera position.
* @Image-2, @Image-3, @Image-4: Lucia's appearance. Preserve her identity, face, skin tone, body proportions, and clothing.
* @Image-5: Seated Lucia, three-quarter side camera angle.
* @Image-6, @Image-7, @Image-8: Exact hairstyles in chronological order.
### SCENE 1 — OCEAN DRIVE | 00:00–00:05
The video starts with Lucia standing right in front of the beauty salon entrance, as shown in @Image-1. She immediately pushes the door forward, and it swings open away from her. She walks inside without stopping.
The salon is located on a lively coastal boulevard in a fictional American metropolis resembling Miami's Ocean Drive in Vice City. The ocean is on the LEFT, while colorful shops, boutiques, and storefronts line the RIGHT side.
Palm trees, pastel Art Deco buildings, pedestrians, and moving traffic create a vibrant open-world atmosphere. The camera follows Lucia from behind at a natural third-person gameplay distance.
Her movement is natural and player-controlled, like authentic GTA VI gameplay.
### SCENE 2 — ENTERING THE SALON | 00:05–00:09
Lucia pushes the door open and walks inside. The third-person camera follows her through the entrance.
Inside is a typical American hair salon. Large mirrors and hairdressing stations are arranged in a row along the LEFT wall. Several identical styling chairs stand in front of the mirrors. A female customer sits in one chair while a Black female hairstylist actively cuts her hair.
The empty styling chair Lucia will use is positioned directly in front of one of the mirrors along the wall, alongside the other styling stations. It is NOT in the center of the room and does NOT stand apart from the other chairs. It is an ordinary salon chair, identical in design to the others, distinguished only by a PINK NEON LIGHT OUTLINE surrounding it. The pink glow is part of the physical environment, NOT a UI marker or overlay.
Lucia walks directly toward this highlighted chair.
**STRICT TIMING: Lucia must enter the salon and reach the chair within 10 seconds maximum. No unnecessary pauses or wandering.**
### SCENE 3 — SEATED VIEW | 00:09–00:12
Cut to a closer three-quarter side view matching @Image-5. Lucia is seated in the highlighted styling chair in front of the mirror along the wall, among the other styling stations. The chair is identical to the surrounding salon chairs. Her face and hair are clearly visible.
Keep the camera stable, with the salon visible in the background. Lucia's appearance, clothing, pose, and position remain consistent.
Her hair initially retains its original hairstyle.
### SCENE 4 — HAIRSTYLE CHANGES | 00:12–00:27
Lucia remains seated in the same position. Maintain the same camera angle, framing, lighting, and character placement throughout all three changes.
**00:12–00:17 — Hairstyle 1:** Her hair instantly switches to the exact hairstyle in @Image-6.
**00:17–00:22 — Hairstyle 2:** Her hair instantly switches to the exact hairstyle in @Image-7.
**00:22–00:27 — Hairstyle 3:** Her hair instantly switches to the exact hairstyle in @Image-8.
Each change is abrupt, occurring in a single frame or almost instantaneously, like selecting a hairstyle in a video game. No gradual transformation, morphing, stretching hair, magical particles, or visual effects. The previous hairstyle is immediately replaced by the next.
Play a short, subtle, recognizable video-game UI confirmation sound at the exact moment of each change. **Audio only: absolutely no visible menu, cursor, icon, text, HUD, or interface.**
Only Lucia's hairstyle changes. Her face, head shape, clothing, body, and pose remain unchanged.
**IMPORTANT: The world NEVER freezes during hairstyle changes.** The hairstylist continues cutting the other customer's hair, background characters keep moving, and the salon remains active. Outside traffic and interior ambience continue uninterrupted.
### SCENE 5 — FINAL HOLD | 00:27–00:30
Hold the same three-quarter side gameplay view. Lucia remains seated with the final hairstyle from @Image-8, clearly visible.
The salon continues operating naturally. Environmental audio continues.
End directly on the gameplay image. No fade to black, title card, credits, or end screen.
### CAMERA AND GRAPHICS
Authentic third-person GTA VI-inspired gameplay, not a cinematic trailer. Real-time AAA game graphics, detailed 3D models, realistic hair rendering, natural animations, dynamic lighting, reflections, shadows, and richly detailed environments.
Use a natural third-person follow camera outside and a stable, closer gameplay camera inside. Avoid dramatic camera movements, excessive depth of field, slow motion, or elaborate transitions.
Maintain consistent character identity, environment, lighting, and spatial continuity.
### AUDIO
Exterior: passing cars, pedestrians, footsteps, and subtle coastal ambience.
Interior: hair clippers, scissors, quiet conversations, footsteps, and traffic faintly audible outside.
Each hairstyle change has a short, subtle game-selection confirmation sound. Environmental audio continues throughout without interruption. No music is necessary.
### FINAL NON-NEGOTIABLE REQUIREMENTS
1. ZERO visible HUD/UI elements throughout the entire video.
2. Start exactly with @Image-1, with Lucia already standing at the salon entrance and immediately pushing the door open.
3. Preserve Lucia's appearance using @Image-2, @Image-3, and @Image-4.
4. Reach the salon chair within 10 seconds maximum.
5. The selected chair is against the wall in front of a mirror, alongside the other styling stations. It is identical to the other chairs and is NOT in the center of the room.
6. Use @Image-5 for the seated composition.
7. Change hairstyles in the exact order @Image-6 → @Image-7 → @Image-8.
8. Hairstyle changes are instantaneous and accompanied by audio only.
9. The surrounding world remains active during all changes.
10. No camera or scene changes between hairstyle states.
11. No visible interface, text, subtitles, watermarks, or UI at any point.
12. End on the final hairstyle with uninterrupted gameplay footage.
A cinematic live circus performance inside a massive dark arena packed with an excited audience. A highly skilled female acrobat with long blonde hair, wearing a vivid red athletic costume and white shoes, stands on a tall metal platform beside a huge circular ring engulfed in roaring flames. Warm firelight illuminates her face and body against the deep black background.
She takes a moment, gathers momentum, then launches herself dramatically through the center of the flaming ring. Capture her in a powerful backward aerial flip, body fully extended and rotating gracefully in mid-air. As she exits the ring, a majestic white horse gallops beneath her, wearing a dark navy saddle and bridle. She completes the rotation and lands safely onto the moving horse in one fluid, perfectly timed motion.
Camera: vertical 9:16, cinematic telephoto lens, dynamic tracking, slow-motion during the jump, subtle camera shake from the live performance, smooth motion tracking.
Lighting: intense orange firelight, dramatic spotlights, deep shadows, realistic flame illumination, atmospheric smoke and sparks.
Style: photorealistic, high-end cinematic sports footage, realistic anatomy and physics, detailed horse movement, natural crowd reactions, dramatic contrast, shallow depth of field, 4K, ultra-detailed.
Mood: thrilling, dangerous, spectacular, heart-pounding live circus spectacle.
Generate a 30-second photorealistic supernatural comedy in 16:9landscape format, beginning from the supplied opening frame.
0–4 SECONDS — THE DISTURBANCE:
Continue directly from the opening frame. Dax shifts his sneakers on Lena’s seat back, causing a small jolt. Lena closes her eyes briefly, then opens them with restrained irritation.
Evan notices the shoes and turns slightly toward Dax. Keep the popcorn bucket upright on his lap.
4–8 SECONDS — THE WARNING:
Cut to a side three-quarter view showing the two rows. Evan turns back and quietly says, “Feet down.”
Dax responds with a dismissive half-smile and keeps his shoes in place.
Lena remains still, then slowly directs her attention toward him.
8–12 SECONDS — THE LIFT:
Cut to Lena’s face. Her irises develop a faint violet tint. She raises one hand beside her shoulder and turns her palm upward.
Cut wider. Dax’s shoes lift away from the seat back as his knees bend naturally. His torso rises from his chair, followed by his hips and legs, forming one continuous, physically coherent movement.
He looks down in surprise as he floats just above his seat. Nearby patrons glance upward.
12–17 SECONDS — THE LAUNCH:
Cut to a reverse wide shot from behind the audience, establishing the large cinema screen ahead. It shows a dusty, damaged city street with distant zombie-like figures.
Dax rises clear of the seat backs and turns to face the screen. Lena makes one sudden forward pushing gesture.
He accelerates straight toward the screen above the center aisle, without striking any patrons. His arms reach forward instinctively while his jacket and hair trail behind him.
Follow his movement with a controlled tracking shot. Keep the trajectory continuous and clearly directed toward the screen. Normal speed throughout.
17–20 SECONDS — THROUGH THE SCREEN:
Dax reaches the center of the screen. The projected surface ripples inward like a luminous membrane as he passes through it: head and shoulders, torso, then legs and sneakers.
The screen closes behind him. Cut on the same forward movement into the movie world.
He lands on his palms and knees on the dusty street, absorbs the movement with bent elbows, and immediately begins pushing himself upright.
20–25 SECONDS — INSIDE THE MOVIE:
Use a low medium-wide shot. Dax stands, looks behind him, and sees three zombie-like figures approaching. They have dusty torn clothing, pale faces, and stiff movements without graphic wounds.
He turns and runs down the street, glancing back in comic panic.
Briefly show the cinema screen from the audience’s perspective: Dax is now visibly running inside the projected film. This confirms where he has gone.
25–30 SECONDS — THE PAYOFF:
Return to Lena and Evan in their original seats. Lena’s eyes are dark brown again.
Evan looks at her with astonishment, then gently offers the popcorn bucket.
Lena takes one small pinch of popcorn with her right hand, raises it to her mouth, places it inside, withdraws her fingers, and chews naturally. The popcorn disappears only after entering her mouth.
She looks forward with a faint satisfied smile. Evan lets out a small amused breath. Hold their quiet two-shot as light from the movie flickers across their faces.
PHOTOGRAPHY:
Realistic cinema lighting, deep-red seats, cool screen illumination, natural skin texture, and readable facial expressions. Inside the movie, use dusty daylight, desaturated architecture, and grounded shadows.
Use clear close-ups, side coverage, and establishing wides. Match action and travel direction across cuts. Keep the supernatural event fast and decisive.
AUDIO:
Natural theater ambience, seat movement, Evan’s brief warning, a subtle supernatural hum, a sharp rushing sound during the launch, movie-world footsteps, and a quiet popcorn crunch at the end.
Use the provided Mac desktop image featuring Rem and Ram as the first frame. Create a 24-second, 16:9 video with a locked-off camera. Preserve the original anime art style: Ram has pink hair, and Rem has blue hair. Keep their faces, hair accessories, and body proportions consistent throughout.
The two characters remain primarily on the right side of the frame, interacting with the viewer beyond the screen. Keep the file icons on the left, the top menu bar, and the bottom Dock stable. Maintain a gentle, intimate atmosphere with natural breathing, blinking, and subtle fabric movement.
0–8 SECONDS | WELCOME HOME AFTER WORK
Rem leans slightly closer to the screen, looks directly at the viewer, and smiles warmly. In a soft, natural Japanese voice, she says:
「ご主人様、今日も一日お疲れ様でした。」
Ram gently tugs Rem's sleeve. They exchange a smile, then turn lightly together. Their swirling skirts sweep up into an arc of pink and blue fabric, briefly concealing their bodies. Switch their outfits instantly during this occlusion. As the fabric falls, they are wearing loose, comfortable loungewear: pale blue for Rem and pale pink for Ram.
At the same time, the wallpaper shifts to soft pink and purple evening tones. Both smile and gently gesture for the viewer to relax.
8–16 SECONDS | OFFERING A WARM DRINK
Rem lifts a pale blue ceramic cup from below the frame, holding its handle with one hand and supporting its base with the other. She slowly offers it toward the viewer beyond the screen. The cup moves naturally closer to the camera as she extends her arms, without covering her face.
Beside her, Ram looks at the viewer and says in a calm, gentle Japanese voice:
「お仕事は、今日はここまでにしましょう。」
Thin wisps of steam rise from the cup, briefly forming a small heart before dispersing naturally. Rem holds the offering pose for a moment, then gently draws the cup back and sets it down below the frame.
16–24 SECONDS | TIME TO SAY GOOD NIGHT
The wallpaper lighting gradually dims. The clock in the top menu bar changes to “23:48,” suggesting that it is now late at night. All other menu text remains unchanged.
Both characters first glance up at the clock, then look back at the viewer. Ram gently folds her arms and speaks in Japanese with mild exasperation and affectionate concern:
「まだ寝ないの？ もう遅いわよ。」
Rem smiles and reaches back to pinch the edge of the wallpaper behind her. As if drawing a soft curtain, she pulls the evening background into a deep blue night scene. The moving curtain edge passes across their bodies in sequence. Switch each character's outfit instantly while she is concealed: Rem changes into pale blue long-sleeved pajamas, and Ram into pale pink long-sleeved pajamas.
Finally, they settle close together again. Rem gives a small wave, while Ram wears a subtle smile. Their hair tips and cuffs sway gently. Hold this tender final moment for one second.
AUDIO AND VISUAL REQUIREMENTS
Use natural Japanese dialogue throughout. No Chinese dubbing and no subtitles. Rem's voice is soft and warm; Ram's is slightly more composed, with understated care. Synchronize lip movements with the Japanese lines. Only one character speaks at a time, while the other reacts naturally.
Use gentle piano music with a few delicate bell tones. As night falls, the music becomes progressively more soothing. Keep it quieter than the dialogue at all times. Add subtle fabric rustles and the sound of the cup being set down.
Do not zoom, dolly, pan, or tilt the camera. The characters approach the viewer only through their own movements.
Both outfit changes must be instantaneous switches hidden behind fabric occlusion. Clothing must not melt or morph, and the act of changing clothes must never be shown.
All background transformations occur exclusively within the wallpaper layer. The file icons, menu bar, and Dock remain fixed and do not move with the curtain.
Create a 10-second horizontal 16:9 video using the provided Beatrice macOS desktop image as the exact first frame.
Scene and character:
Fixed front-facing camera, one continuous shot, no camera movement or cuts. Preserve the rose-pink and champagne wallpaper, English macOS menu bar, left-side application grid, and bottom Dock.
Preserve Beatrice’s blonde spiral twin curls, pink bows, tiny crown accessory, distinctive eyes, elaborate burgundy-pink dress, and original youthful anime proportions.
Personality:
Beatrice is unmistakably tsundere: proud, easily embarrassed, and defensive about showing vulnerability. She accidentally makes a mess, tries to deny her embarrassment, then fixes everything while acting as though she is merely doing the viewer a favor. Express this through averted eyes, flushed cheeks, a stubborn pout, and precise, slightly impatient gestures.
Timeline:
0–1.5s:
Beatrice stands on the right in her original pose. Her nose twitches. She frowns and stubbornly tries to hold back a sneeze, pressing her lips together and lifting her chin.
1.5–2.5s:
A sudden cute sneeze escapes. Her shoulders bounce and her spiral curls spring. The gust knocks Gmail, Discord, and Microsoft Teams from the rightmost icon column. They tumble down separately and land just above the Dock. Their original slots remain empty.
2.5–3.5s:
She freezes, staring at the fallen icons. Then she notices the viewer looking. Her cheeks turn pink; she immediately turns her face slightly away, puffs her cheeks, and gives the viewer a brief defensive sideways glance.
3.5–5s:
Still pretending to be unconcerned, she lifts one hand and flicks her index finger. A small violet glow surrounds Gmail alone. It rises from the lower desktop and visibly travels back into its original slot.
5–6.3s:
She makes a second brisk gesture. Discord lifts and returns to its own position. She keeps her chin raised but secretly follows the icon with her eyes to make sure it lands correctly.
6.3–7.5s:
A third gesture returns Microsoft Teams. It lands slightly crooked. Her brow twitches; she immediately corrects it with one tiny, impatient finger movement. All three icons are now neatly restored.
7.5–10s:
Beatrice folds her arms, turns her face away with a blush, and says in a small, defensive Japanese voice:
「べ、別に、あなたのためじゃないのよ。」
She briefly peeks back at the viewer from the corner of her eyes, then looks away again with a stubborn pout. Her curls gently settle. No friendly wave or openly sweet smile.
Audio:
Playful pizzicato and light harpsichord, a small sneeze, soft icon landing taps, and three delicate magical chimes. Deliver the Japanese line quickly but clearly, with a slight stumble on the first syllable and embarrassed emphasis. Match her lip movements to the dialogue. No subtitles.
Constraints:
Restore the three icons strictly one at a time. Each icon must visibly travel from its landing spot to its original empty slot before the next moves. Keep magical effects small and controlled. No teleportation, duplication, or simultaneous restoration.
All other desktop icons, the menu bar, Dock, and wallpaper remain stationary. Preserve her original character design, clothing, and anatomy. The final emotional beat is concealed embarrassment behind a proud attitude.
Pollo AI Wan 3.0｜雫 VS 巨大妖怪 ニコガサネ｜30秒ワンカット
Create a 30-second, 16:9, full-screen cinematic Japanese anime battle with synchronized Japanese dialogue, sound effects, music, and environmental audio.
The physical world camera remains ONE CONTINUOUS UNBROKEN SHOT. No hard cuts, montage, split screen, or teleporting camera. The camera continuously dollies, circles, dives, cranes, and passes through debris, but it must not become a third-person game camera behind Shizuku.
Only during the final iai preparation, allow one 0.20-second eye-and-hand anime cut-in overlay. The underlying world camera keeps moving.
Visual style: theatrical hand-drawn Japanese anime, painted sunset background, selective animation, missing in-betweens, extreme smear frames, readable key poses, strong impact deformation, black ink speed strokes, vivid sword effects, and heavy causal building destruction. Avoid smooth 3D-game motion.
REFERENCE ROLES
Reference 1 controls Shizuku’s identity. Preserve her short black bob, face, blue-white-red oversized jacket, gray hood, black loose pants, blue chunky sneakers, patched black backpack with plush toys, and approximately 1.6 m height.
Ignore any sword placement shown on her back in the reference. The prompt’s single left-waist weapon placement has priority.
Reference 2 controls Nikogasane’s identity and scale. Nikogasane is approximately 35 m tall: pale blue-green round body, cream belly, two cat-like ears, exactly two round eyes, one flat red smiling mouth, two thick arms, multiple short feet, and heavy soft bumpy skin. Its head towers above two-story buildings and Shizuku is smaller than one finger.
Reference 3 controls the initially intact sunset town, intersection, shops, houses, poles, wires, weeds, Mount Fuji, orange-pink clouds, cyan sky, and long shadows.
Reference 4 controls only the exact title composition and typography:
「雫 VS 巨大妖怪 ニコガサネ」
Treat the existing title as pre-composed artwork. Do not redraw, translate, shorten, replace, or misspell it.
Reference 5 controls only the final low iai posture: deep crouch, front knee bent, rear knee near the ground, hips low, torso leaning forward, right hand on hilt, left hand pulling the scabbard backward, blade fully sheathed, body compressed like a spring.
Shizuku’s identity and clothing remain from Reference 1.
ABSOLUTE SINGLE-WEAPON LOCK
Exactly one katana blade and one matching scabbard exist. They are one weapon assembly.
The scabbard remains fixed at Shizuku’s LEFT WAIST.
No sword, hilt, or scabbard on her back or backpack.
No second weapon.
No dual wielding.
The scabbard is never removed and never used to block.
Before the first draw, the blade is fully sheathed.
After drawing, exactly one blade is in her right hand and the empty scabbard remains at her left waist.
During final sheathing, the same blade enters the same scabbard and becomes fully hidden.
MONSTER ATTACK READABILITY — ABSOLUTE OVERRIDE
Nikogasane’s mouth is NOT an attack organ.
Its mouth remains a FLAT, CLOSED RED SMILE throughout the battle.
It never opens.
No teeth.
No tongue.
No visible mouth interior.
No vortex.
No suction.
No projectile.
No beam.
No hand, arm, tentacle, or object may emerge from its mouth.
Every attack originates visibly from one of its two shoulders and arms.
For every arm attack, show the complete connected silhouette in the same shot:
shoulder rotates → elbow bends or extends → forearm follows → hand reaches the target.
The shoulder, elbow, forearm, and hand must remain visibly connected to the side of the torso.
Never align an attacking arm with the mouth.
Never frame a hand as though it emerged from the mouth.
Use only these clear monster attacks:
right-arm horizontal backhand sweep
left-forearm block followed by right open-palm backhand
both-hands ground slam
No mouth attack.
No additional special ability.
CAMERA AND FACE RULES
Show Shizuku’s face clearly at least five times:
opening cigarette close-up
confrontation line
first clash
damaged close-up
final iai close-up
During combat, the camera mainly travels across her front hemisphere:
left profile → front three-quarter → right profile.
Rear views last no longer than 0.30 seconds.
Never chase directly behind her.
Never center the backpack during an important action.
For enemy strikes, use a three-quarter side wide angle showing:
Nikogasane’s shoulder
elbow
forearm
hand
Shizuku
the target building
all in one readable composition.
When Shizuku is knocked backward, the camera retreats AHEAD of her while facing her, then enters the building before she crashes through the wall.
For the final iai, use a low lateral side-wide camera approximately 90 degrees to her travel direction. The start position, Nikogasane, and the empty distance between them must be visible.
HIGH-SPEED ANIMATION GRAMMAR
Every Shizuku dash, dodge, or reposition finishes within 0.25–0.50 seconds.
Never show a normal running cycle.
For each movement show only:
compressed anticipation for 2–3 frames
one stretched smear or black-blue silhouette for 1–2 frames
omit all other intermediate body positions
immediate arrival, contact, defense, recoil, or landing
The camera may lose her for 0.10–0.18 seconds and whip to recover her.
Dust and debris react after she has already arrived.
Use only 2–3 frames of hit-stop.
SHORT JAPANESE DIALOGUE
Generate exactly three lines from Shizuku in the same low, dry, slightly husky, restrained young Japanese female voice:
「……ふざけんな。」
「痛ぇな。」
「消えろ。」
No other words.
No repetition.
No subtitles.
No narration.
No translation.
Keep her mouth visible and reduce music while she speaks.
30-SECOND CONTINUOUS TIMELINE
0.00–1.50 — Smoking face close-up
Open directly on Shizuku’s left-front three-quarter face.
Her face fills about 40% of frame.
She crouches beside the intersection with a cigarette at her mouth.
The single katana is fully sheathed at her LEFT WAIST.
No weapon is on her back.
Push toward her eyes while drifting 30 degrees across her face.
She inhales.
The ember glows.
She exhales thin smoke.
Hair, jacket, strap, and plush toys move in the wind.
Nikogasane rises beyond the rooftops.
Audio:
wind
wires
cigarette
cloth
distant giant footstep
No music.
1.50–3.40 — Stand, scale reveal, first line
Shizuku stands.
Arc from her left profile through a frontal three-quarter angle, keeping her face visible.
Nikogasane steps into the intersection at true 35 m scale.
Its flat red smile remains closed.
One arm brushes a roof.
Tiles slide and wires stretch.
Shizuku looks up and says exactly:
「……ふざけんな。」
3.40–4.30 — Exact title
Match Reference 4 and display exactly:
「雫 VS 巨大妖怪 ニコガサネ」
Hold 0.85–0.90 seconds.
Audio:
one taiko hit
metallic resonance
short sub-bass pulse
No speech.
At 4.30 seconds, Nikogasane begins the first attack.
4.30–7.20 — Clear right-arm sweep and omitted dodge
Use a three-quarter side wide shot.
Nikogasane pulls its RIGHT shoulder and elbow backward in a readable anticipation pose.
Its mouth stays closed.
The shoulder rotates.
The connected right forearm and open hand sweep horizontally across the street.
Shizuku compresses low and disappears 14 m sideways in 0.35 seconds.
Show only:
starting pose
one black-blue smear
delayed dust
immediate low landing
No running.
The right hand misses her and physically tears through three storefronts:
hand contact → wall fractures → beams snap → glass bursts → roofs fall a fraction later.
The camera passes through debris and circles toward Shizuku’s front-side face.
7.20–10.80 — First clash and clearly connected counterattack
Shizuku draws the single katana once and closes 12 m in 0.4 seconds.
She slashes Nikogasane’s LEFT forearm.
The left forearm remains visibly connected to the left shoulder and blocks the blade by compressing.
The blade visibly stops with:
white-blue light
orange sparks
pressure ring
2–3 frames of hit-stop
Orbit across Shizuku’s front hemisphere so her focused face is visible.
Nikogasane releases the blade.
It visibly winds up its RIGHT shoulder.
The right elbow bends.
It performs one connected open-palm backhand toward Shizuku’s torso.
Keep the complete shoulder-to-hand chain readable.
Mouth closed.
10.80–15.50 — Heavy hit into building; preserve this impact
The right open palm lands through Shizuku’s partial single-blade guard.
She does not fully evade.
Show:
palm contact
painful face reaction
body folding from impact
2-frame impact flash
one long backward smear
Shizuku is launched about 18 m.
The cigarette flies past the lens and dies on the road.
The camera retreats in front of her while facing her.
It enters an intact shop first and sees her body crash through the exterior wall toward the camera.
Show wood, plaster, glass, shelves, and dust breaking inward around her silhouette.
She strikes an interior pillar.
She falls to one knee.
She remains down briefly.
Arc through dust into a close front three-quarter face shot.
She says exactly:
「痛ぇな。」
This building-impact phase is essential.
Do not shorten or remove it.
15.50–20.60 — Recovery and one compact counter exchange
Shizuku plants one sneaker against the broken pillar and rebounds outside in 0.4 seconds:
compressed pose
one smear
pillar snapping after she leaves
immediate exterior landing
The roof collapses behind her.
She closes 15 m in one omitted burst and attacks the monster’s left arm.
Use exactly three readable beats.
Beat 1
Shizuku performs one horizontal slash.
Nikogasane blocks with the connected left forearm.
The blade stops.
Beat 2
Nikogasane swings its connected right forearm downward.
Shizuku slips to the side in one 0.30-second smear.
The hand destroys a shop façade.
Beat 3
Shizuku pivots from the dodge and lands one short cheek slash.
Cyan mist and orange sparks erupt.
The mouth remains closed and produces no attack.
No extra combo.
20.60–23.30 — Two-hand ground slam creates iai distance
Pull wide to show both fighters.
Nikogasane raises BOTH connected arms high above its head.
Show both shoulders, elbows, forearms, and hands clearly.
Both open palms slam into the intersection.
Create:
one circular shockwave
cracked asphalt
rising road slabs
bending poles
breaking windows
collapse of already-damaged storefronts
Shizuku meets the first pressure edge with the single blade, but the shockwave drives her backward.
Show:
one compressed defensive pose
one 0.35-second backward smear
immediate landing 18 m away
She lands on stable cracked asphalt.
23.30–25.35 — Complete sheathing, low iai, final line
Use a low front-side orbit.
Never move behind her.
Shizuku performs the complete sheathing:
blade tip enters the single scabbard → blade slides fully inside → guard reaches the scabbard mouth → clear metallic click.
After the click, no exposed blade remains.
She instantly drops into the low spring-loaded Reference 5 posture:
front knee deeply bent
rear knee close to road
hips extremely low
torso leaning forward
right hand on hilt
left hand pulling scabbard backward
blade fully hidden
eyes locked on Nikogasane
Move close to her face and hands.
She says exactly:
「消えろ。」
On the final syllable, show the only 0.20-second cut-in:
both eyes
right fingers tightening
left thumb cracking the guard open
thin white-blue glint
black ink diagonal
tiny orange sparks
25.35–25.65 — INSTANTANEOUS IAI CUT-THROUGH
The distance closure and the sword slash are ONE INDIVISIBLE ACTION.
They occur at the exact same instant.
Do not show Shizuku approaching before the slash.
Do not show running.
Do not show stepping.
Do not show jogging.
Do not show repeated leg movement.
Do not show a separate dash followed by a slash.
Use a low lateral side-wide camera.
Continue a slight sideways camera slide, but do not orbit behind her.
Show only THREE VISUAL STATES.
STATE A — 2–3 frames
Shizuku remains in the extremely low, fully sheathed iai pose, 18 m in front of Nikogasane.
STATE B — 1–2 frames
Her rear foot explodes against the asphalt.
The blade clears the scabbard at the exact same instant that her entire body becomes one stretched black-blue silhouette.
A brilliant horizontal white-blue blade streak physically passes through Nikogasane’s torso.
The sword light stays attached to the real moving blade.
The light fills the lens for a 2-frame flash, hiding all intermediate travel.
STATE C — immediate next readable pose
Shizuku is already 8–10 m BEHIND Nikogasane in a low draw-through finish pose.
single katana fully drawn
front knee bent
rear leg extended
torso low
body facing away from the monster
There are ZERO readable body positions between the starting pose and finishing pose.
The slash and spatial displacement are simultaneous.
It must feel like she vanished, cut completely through the monster, and appeared behind it in one instantaneous motion.
No stationary slash.
No energy beam.
No run-up.
No jump.
No air time.
No pursuit camera.
25.65–30.00 — Clear physical defeat motion
After the cut, Nikogasane cannot attack again.
It does not instantly dissolve.
25.65–26.05
Near silence.
Shizuku holds the low finish behind it.
Nikogasane freezes.
A thin horizontal white-blue cut line appears across its torso.
26.05–26.80
The monster’s eyes lose focus.
Its flat smile distorts.
Both connected arms drop.
Its upper body buckles.
Its multiple feet fail.
26.80–27.80
Nikogasane staggers backward with TWO heavy steps.
Each step crushes the road.
Its back hits two already-damaged buildings:
body contact → walls cave inward → columns break → roofs collapse → wires snap.
27.80–28.70
Its feet buckle completely.
The 35 m body falls backward and sideways onto the street.
Create:
massive ground impact
rolling dust wave
collapsing nearby roofs
shaking utility poles
Keep the body fully visible.
28.70–30.00
Only after the physical fall, white-blue cracks spread from the cut line.
Nikogasane gradually dissolves into:
cyan mist
soft glowing fragments
supernatural particles
Curve the camera from Shizuku’s side toward her front three-quarter face.
End with:
Shizuku holding the single drawn katana
fallen and dissolving Nikogasane
collapsed buildings
Mount Fuji
orange sunset
AUDIO
Opening:
wind
wires
cigarette
cloth
distant footstep
Battle:
giant arm displacement
elastic body impacts
initial katana draw
blade resistance
timber
concrete
glass
wires
hybrid taiko
distorted percussion
Building hit:
heavy open-palm impact
air pressure
wall rupture
interior crash
debris settling
Iai preparation:
music strips down
blade sliding into scabbard
metallic click
short breath
「消えろ。」
0.10-second silence
Iai:
explosive rear-foot launch and draw sound occurring as one sound
huge passing slash
sub-bass pressure wave
Defeat:
two heavy backward steps
building collision
structural collapse
full-body impact
delayed supernatural dissolution
ABSOLUTE NEGATIVE CONSTRAINTS
No mouth opening.
No teeth.
No tongue.
No mouth hand.
No arm emerging from mouth.
No mouth projectile, vortex, suction, or beam.
No disconnected limb.
No attack whose shoulder and elbow are hidden.
No unclear monster attack.
No riding, standing, or running on Nikogasane’s arms, head, mouth, or body.
No second sword, blade, hilt, or scabbard.
No weapon on back or backpack.
No scabbard used as a weapon.
No normal running.
No slow jogging.
No repeated steps.
No complete travel animation.
No static game camera.
No prolonged rear view.
No hidden face during dialogue.
No long dialogue.
No extra words.
No repeated speech.
No subtitles.
No removal of the building-impact damage phase.
No skipped sheathing.
No upright iai stance.
No exposed blade before the final draw.
No run-up.
No jump.
No airborne finisher.
No stationary slash.
No detached slash beam.
No separate dash and slash.
No intermediate body positions during the final iai.
No instant monster disappearance.
No mist-only defeat.
No dissolution before staggering, building impact, and full-body collapse.
No extra characters.
No duplicate Shizuku.
No shrinking Nikogasane.
No title misspelling.
No logo.
No watermark.
No blood, gore, or dismemberment.
The final result must clearly show:
readable arm attacks from the shoulders → heavy hit into the building → recovery → complete sheathing → low iai pose → simultaneous instantaneous distance closure and one slash → Shizuku already behind the monster → clear stagger, building impact, fall, and delayed dissolution.
15-second realistic cinematic horror sequence, shot on 35mm film, grey overcast beach at dusk, cold blue-grey colors, wet sand, subtle film grain. Exactly one person: a woman in a yellow raincoat standing on the shore.
[0-5s] Wide shot of a calm beach. The woman stands at the water's edge looking out at the sea. Suddenly the water starts pulling back fast, far faster than any normal tide, exposing wet sand and rocks for hundreds of meters.
[5-10s] The viewer rises high above her. As the sea drains away, the exposed seabed reveals rows and rows of pale human figures standing perfectly still on the sand, dozens of them, all facing the ocean, their backs to the woman.
[10-15s] Close-up on the woman's face as she steps back in horror. Cut to wide shot: all the figures on the seabed turn around at the exact same moment and face the viewer. In the far distance, a massive wave is rising.
Negative prompt: cartoon, anime, 3D render, sunny weather, bright colors, swimmers, figures moving out of sync, static scene, nothing happening, distorted faces, extra limbs, morphing, flickering, slow motion, watermark
Use the uploaded image as the exact source image and first frame. Preserve the original character design, composition, face, pale skin, glowing icy blue eyes, braided black hair, cybernetic head implants, exposed mechanical neck and chest structure, white and cobalt-blue jacket, blue serpentine mechanical creatures, labels, symbols, and the clean white background. Keep the result extremely faithful to the original illustration. The animation should feel like the original artwork has come alive, not been redesigned. Create a premium, stylish, high-end anime animation with a refined cyberpunk aesthetic. The overall impression should be elegant, cold, hypnotic, dangerous, and visually luxurious. Emphasize precise Japanese anime craftsmanship: crisp linework, beautiful cel shading, highly detailed mechanical rendering, controlled highlights, graceful motion, and strong temporal consistency. The animation should look polished, expensive, and intentional. The central character remains calm, dominant, and nearly still. She faces forward and maintains an unwavering, piercing gaze directly into the camera. Her expression is cool, emotionless, intelligent, and slightly predatory. Avoid exaggerated acting. Only subtle micro-movements are allowed in her eyelids, breathing, neck posture, and facial tension. Her presence should feel powerful even in stillness. The main animated elements are the multiple blue mechanical serpents surrounding her. They move with smooth, layered, snake-like elegance. Each serpent has slightly different timing, path, and rhythm, so the movement feels organic and sophisticated rather than synchronized. Their bodies undulate in natural S-curves with believable weight, inertia, and continuous motion. Some serpents slowly raise their heads, some glide across her shoulders and around her face, some tighten or loosen their coils, and some subtly cross behind or in front of the composition to reshape the frame. Their movement should feel alive, hypnotic, and controlled, never frantic. Preserve the segmentation, scale texture, metallic fittings, joints, and cable-like details. Small articulated components should shift with delicate mechanical precision. Her long green tongue moves like a real serpent tongue but remains elegant and stylized. It slowly extends, retracts, and sways with controlled, organic softness. The forked tip performs subtle tasting-the-air flicks. The tongue should feel eerie, sensual, and mesmerizing, never comedic. The tongue motion should complement the serpents’ motion, acting like a focal accent that draws attention toward her mouth and face. Add subtle secondary motion throughout the scene. Her shoulders and upper torso rise and fall gently with slow breathing. Tiny motions should ripple through the hood, jacket folds, loose blue fabric, fine cables, and a few hair details. The exposed mechanical chest and neck assembly should feel alive through micro-movements: slight cable tension shifts, minute piston adjustments, tiny joint rotations, and restrained vibration. These details must remain subtle and tasteful, never distracting from the face and serpents. Lighting should remain clean and striking. Preserve the white background and the strong graphic contrast between the white, cobalt blue, black mechanical detailing, and neon green tongue. Let reflections glide softly across the blue serpents’ scales and metallic joints. Metallic parts in the neck and chest may shimmer subtly. The green tongue should remain a vivid accent with slight wet gloss or faint luminous emphasis. Maintain a polished illustration-like finish while introducing dimensionality through moving highlights and soft shadow transitions. The camera should perform an extremely slow and smooth frontal dolly-in over the full 15 seconds. The framing begins as a stable medium close-up and gradually becomes slightly tighter and more intense. The camera should never move abruptly. No shake, no tilt, no sudden zoom, no spinning, and no dramatic reframing. The face, eyes, and tongue should become increasingly dominant as the sequence progresses, while the surrounding serpents continue to frame the composition. The movement of the camera should feel cinematic and hypnotic, reinforcing tension through slow approach. Structure the motion over the 15-second duration with deliberate pacing: Seconds 0–3: begin in near stillness. Only the faintest breathing, tiny hood and cable motion, and a barely perceptible serpent twitch. Establish tension and the character’s unwavering blue-eyed stare. Seconds 3–6: the serpents begin to awaken. One or two serpents slowly lift their heads and start gentle undulating motion. Reflections start traveling across their blue surfaces. The tongue begins a slight side-to-side sway. Seconds 6–9: more serpents join the motion. The body language of the serpents becomes richer and more layered, circling her shoulders and face with elegant S-curve motion. The tongue extends a little farther and retracts slowly, with subtle fork-tip flicks. The camera continues its slow push inward. Seconds 9–12: reach the most hypnotic phase. The serpents are now fully active but still graceful and controlled, framing her face dynamically. Their movement should feel luxurious and predatory. The tongue performs its most visually striking serpent-like motion here, synchronized with the heightened tension. Highlights glide across scales and metal details. Her gaze remains locked and emotionally unreadable. Seconds 12–15: sustain and slightly resolve the intensity. The serpents continue moving but begin to feel more settled, as if entering a poised equilibrium. The camera reaches its closest framing, emphasizing her eyes and green tongue while still preserving enough serpents in frame for visual impact. End on a powerful, composed, hypnotic final moment that feels elegant and complete. The emotional tone should be futuristic, eerie, sensual, intelligent, and luxurious. The final result should resemble a high-budget anime opening shot or a premium animated key visual. Prioritize elegance, polish, and controlled visual storytelling over excessive motion. Every motion should feel deliberate and aesthetically composed. Maintain strict model fidelity and visual stability. Do not alter the character’s face shape, eye design, hairstyle, skin tone, clothing, mechanical anatomy, serpent design, or color palette. Do not add new objects, extra serpents, extra tongues, extra limbs, or background elements. Do not morph the face or let the eyes drift. Do not let the serpents fuse together unnaturally or intersect in broken ways. Do not allow clipping through the body or clothing. Do not change the jacket symbols or graphic details. Do not shift into 3D, live action, or a different illustration style. Avoid flicker, line wobble, perspective drift, detail melting, unstable anatomy, scale inconsistency, random deformation, abrupt motion, or composition changes.
15-second, 16:9, cinematic realistic fantasy short developed strictly from <Picture 1>. Lock the whole sequence inside the same small lake boat space only: the same yellow-blue old wooden boat, the same dark lake water, the same orange-cyan umbrella, the same reeds and purple flowers, and the same woman, dog, and book. No second location, no extra people, no black ending. Preserve the exact same adult woman: auburn-brown long hair, straw hat, deep blue floral bikini, same face, same proportions, same seat on the right side of the boat. Preserve the same brown-white beagle with cyan collar on the left half of the boat. Preserve the same gilt-edged dark green hardcover book, starting half-open in her left hand between them. The emotional line is restrained: she wants to keep reading, the dog interrupts, the book reveals something unnatural, and she shuts it decisively before settling into wary tenderness.
Blocking and continuity: left dog, right woman, middle book, rear umbrella remain stable throughout. She never stands. The dog never crosses into her side. The book follows one strict state progression only: half-open in her hand -> dragged down by the dog onto the yellow deck while still open -> closed by her own hand. Camera must always remain on the same visible gunwale side as in <Picture 1> and never cross the axis formed by the woman and the book.
[00:00-00:02.5]
Slow push-in, oblique half-body close shot. She leans from a slightly reclined posture toward the book. Her right palm slides along the damp gunwale to stabilize her center of gravity while her left hand lifts the book and teases up half a page with her thumb. One natural inhale lifts the ribcage and swimsuit slightly, then a soft exhale returns it. Water droplets on the shoulder and neck catch light. Focus settles on the page she is about to turn but has not turned yet. White smoke begins to emerge from the page gutter. Reflections on the water and reeds continue moving gently. Her gaze stays on the text, lips slightly parted from concentration.
[00:02.5-00:03.5]
Pan left along the white smoke toward the dog. The dog’s nose follows the smoke. Its eyes shift to the back of her neck before the head fully turns, ears lifting and hind legs storing energy. The dog’s launch toward the right bridges into the next beat.
[00:03.5-00:05.7]
The dog bites the lower-left corner of the page. The page tears with a short rip and a few paper fibers. A whip-flash shake catches the woman to the right. She recoils right-rear, her left hand losing the book, her right palm slamming down on the gunwale, and the boat tilting then rebounding. Keep face, shoulders, and upper torso together in frame. As the shoulders and back pull away, the chest follows with one short natural inertial lift; when her body stops and the boat rebounds, that motion resolves into two restrained offset responses, one vertical and one slight lateral, both physically damped by the swimsuit and body weight. Hair whips late and a few droplets leave the skin. Use only a very brief local slow-down at the rebound peak, then return to normal speed. By the end of the shot, the book falls toward the center deck, still on the dog’s side of the axis, and the dog remains left of her.
[00:05.7-00:08.4]
Cut to a low camera inside the same visible side of the boat. The frame stabilizes and depth shifts smoothly. The hard corner of the book cover hits the yellow deck with a muted knock; a little dry worn corner debris breaks off; the pages tremble and remain open. Show the book first, then shift focus to the continuous white smoke rising from the same page gutter. The smoke is drawn longer by the breeze under the umbrella, twists, and condenses into a semi-transparent hand reaching toward where her face was a moment earlier. The smoke-hand casts a passing hand-shaped shadow under the umbrella, consistent with the existing light direction. Her right hand is still bracing the boat. Her gaze moves from the dog to the hand-shadow. The dog has retreated back to the left side of the book. Lake ripples are caused only by the rocking of the boat. Purple flowers and reeds move irregularly in the breeze.
[00:08.4-00:11.2]
She does not stand. Her abdomen tightens and her body leans from the right toward the left-front. Her right hand continues to stabilize on the gunwale while her left hand reaches and catches the book cover. The camera tracks left along the same side of the boat, holding her eyes, reaching hand, upper torso, and the book corner in one frame. The forward lean lowers the chest with the shift in center of gravity; the abrupt stop before closing the book creates one short inertial rebound; quick breathing adds a small continuous rise-fall through the chest and swimsuit. Hair tips brush the shoulder and the straw-hat brim trembles lightly. She sets her jaw and slaps the book shut with one decisive movement. The pages clap closed, the cover vibrates, and the smoke-hand collapses from the fingertips backward into fine strands that retract rapidly along the original path back into the book gutter. The umbrella shadow of the hand shrinks out in sync. No floating smoke residue remains.
[00:11.2-00:15.0]
Slow pull-back. The book remains closed on the center deck, and her left hand is still pressing it down. Her upper chest rises and falls through two short recovery breaths, then gradually returns toward slower breathing. The dog approaches from the left and touches the back of her hand with its nose. She lifts her eyes once to confirm there is nothing above under the umbrella, then her left hand leaves the book cover and moves to the dog’s cyan collar, while her right hand still steadies the gunwale. She says softly in English, with breath not fully settled yet: <d>[English] Good dog.</d> The line is short and light. The camera pulls back to hold the half-body relationship of the woman and dog under the same umbrella. The boat rocks gently. The dog’s tail flicks only once. Her left hand finally rests on the dog’s collar. Do not cut to black.
Lighting and color: preserve the original warm-umbrella and cold-lake palette. Warm top light filtered through the umbrella illuminates her skin. The lake gives soft cyan reflections in shadow. Her skin stays naturally warm with real water-droplet sheen. The smoke is cold silver-white and only local. When the smoke-hand appears, only the local light under the umbrella and environmental sound are slightly suppressed; do not change sun direction, time of day, or overall color temperature. After the book is closed, the original warm-cool balance returns.
Physical logic: the old wooden boat tilts and rebounds locally from the dog’s leap and her bracing hand, and water slaps the hull into uneven ripples. The umbrella fabric and ribs tremble with the boat’s motion but do not come loose. Page tearing, cover striking wood, tiny dry debris, palms pressing wood, and droplets leaving wet skin all carry correct force and sound. The smoke originates only from the book pages, only inside the boat and under the umbrella, never passing through skin, and never causing weather or location shifts. Hair moves in layers; the wettest strands lag the most. Upper-body motion is always driven by breathing, torso acceleration, abrupt stopping, and boat rebound, with real support and no exaggeration.
Hard negative constraints: no identity drift, no face deformation, no extra people, no extra limbs, no extra animals, no second location, no wide cutaway, no magical particle storm, no fake smoke residue, no zero-gravity hair, no cloth drift, no black ending, no subtitles, no watermark.
overall_soundscape:
The base ambience is intimate and natural: dark lake water lapping the boat, reeds moving, pages rustling, quiet dog breathing, slight umbrella fabric tremor, and soft wet skin or cloth movement. The opening includes page handling, thumb on paper, a subtle breath in, light contact of her hand on the damp gunwale, and the first faint hiss of white smoke from the book gutter. When the camera pans to the dog, bring forward the dog’s sniffing and body tension. At the grab, layer paw scrape on wood, a short paper tear, her sharp inhale, the right palm striking the gunwale, and the boat’s sudden wood-and-water reaction. Immediately after, reduce the environmental bed so the smoke hiss, slight umbrella-rib vibration, and the low uncanny presence of the smoke-hand dominate the foreground. When she lunges back in and shuts the book, the cover closes with a firm muffled clap that cuts the unnatural sound off. Then the lake sound and normal ambience return. In the ending, her breathing is still slightly unsettled as she says <d>[English] Good dog.</d> softly, and the dog answers only with a small nose touch and one tail flick.
non_diegetic_music:
Keep the score restrained and cinematic. Begin with a soft warm-cool suspended texture under the water and reeds, almost imperceptible, supporting the concentration of reading. As the smoke appears and the dog notices it, introduce a faint low unease without overpowering natural sounds. At the book snatch and recoil, give one brief accent and then immediately thin the score so paper, breath, boat movement, and smoke hiss lead. During the smoke-hand section, hold a narrow suspended tone with minimal low-frequency pressure. When the book is shut, let that tension collapse quickly into a lighter, warmer emotional residue. End with only a small intimate undertone beneath the returning lake ambience, with no melodramatic swell and no hard stop.
Ultra cinematic macro shot: A calm mountain lake at dawn reflects the first golden rays of sunlight. Tiny ripples spread across the mirror-like surface as morning mist drifts gracefully above the water. Billions of sparkling droplets slowly rise into the air, suspended as if gravity has stopped. They spiral together with incredible fluid realism, naturally sculpting the word "MORNING" entirely from crystal-clear water. The liquid letters shimmer with golden reflections before gently collapsing into a spectacular explosion of droplets illuminated by the sunrise. Hollywood title sequence, ultra-realistic fluid simulation, volumetric lighting, IMAX quality.
Slow push-in with slow orbital camera movement. A colossal white tiger lowers its head and slowly extends its right front paw toward a woman in a flowing red robe. Mist swirls around the tiger and drifts across the woman, her red garments billowing in the wind. Ends on a tense close-up standoff between the woman and the white tiger. Cinematic dark fantasy, Chinese temple silhouettes, warm orange lantern glow against cold blue-grey mist, high detail fur. No subtitles, no on-screen text, no watermarks.
Create a 15-second, 16:9 photoreal cinematic sci-fi mystery sequence with native stereo audio.
Use Image 1 as the strict anchor reference for the explorer’s identity, face, short dark hair, charcoal hooded coat, black tactical clothing, boots, gateway design, wet obsidian landscape, jagged rock formations, lighting, materials, and cold blue/crimson color palette. Use Images 2–4 as sequential visual keyframes and preserve complete character, costume, environment, scale, and screen-direction continuity.
[0–3 seconds] Begin with Image 2. Ultra-wide symmetrical establishing shot from behind the explorer. She walks slowly toward the colossal circular gateway, remaining very small against its immense scale. The camera performs a restrained forward dolly at ground level, following the crimson reflection across the wet surface. Her boots create small splashes. Mist drifts laterally around the distant rocks. The glowing ring pulses faintly, but its center remains completely black.
[3–6 seconds] Cut on her next footstep to a low rear three-quarter tracking shot. Move beside her at walking speed while keeping the gateway fixed ahead. Her coat responds naturally to the wind. The crimson illumination travels across the wet fabric and reflective ground. Do not change her walking direction or position relative to the gateway.
[6–9 seconds] Cut to Image 3. Tight three-quarter facial close-up. Slowly push toward her eyes as she stops walking and looks upward. Preserve her exact facial identity. Her expression shifts subtly from focused caution to disturbed recognition. Fine strands of hair move across her face. The gateway remains softly recognizable behind her. A dim crimson reflection appears in her eyes.
[9–12 seconds] Cut to a close shot of her gloved hand rising cautiously into frame. Track forward with the hand as she approaches the threshold. The air near the black center distorts slightly, pulling mist and loose droplets toward it. Keep the effect restrained and physically coherent, with no lightning, explosions, or fantasy energy.
[12–15 seconds] Transition into Image 4. Medium-wide side view near the gateway. She extends her hand toward the pitch-black center but stops just before contact. The crimson ring suddenly dims for half a second. All ambient wind and rain sound disappear. From inside the void, a quiet version of her own breathing answers one beat later. She freezes. End on a slow push toward the narrow space between her fingertips and the darkness, cutting to black before contact.
Visual style: premium live-action science-fiction film, realistic skin and fabric, cold blue ambient light, restrained crimson rim light, deep blacks, wet reflections, fine atmospheric mist, subtle film grain, controlled contrast, 24 FPS, natural motion blur, cinematic depth of field.
Audio: distant low-frequency gateway resonance, cold wind, light rain, boots splashing through shallow water, fabric movement, restrained breathing, and a deep sub-bass pulse that grows gradually. Remove all environmental sound during the final silence, leaving only the delayed breathing from inside the void.
No dialogue, subtitles, title cards, logos, watermarks, extra characters, vehicles, weapons, creatures, lightning, explosions, camera spins, random zooms, rapid montage, costume changes, facial changes, distorted hands, warped gateway geometry, daylight, or bright saturated colors.
[FORMAT]
Create a 10-second, 16:9 photoreal cinematic wide-shot flyby of a stealth aircraft activating optical camouflage at extreme speed.
[IDENTITY]
Preserve a wide tailless delta wing, sharp faceted edges, flat wedge-shaped nose, recessed dark cockpit, matte charcoal surface, twin buried exhausts, no insignia, and rigid non-morphing geometry.
[BEATS]
[0–2 seconds] A vast open sky above distant clouds. The stealth aircraft suddenly enters from the far left and races laterally across the wide frame at extreme speed.
[2–4 seconds] The camera performs a fast, controlled pan to track the aircraft. Strong background parallax and slight motion blur emphasize its velocity while its shape remains readable.
[4–6 seconds] Optical camouflage sweeps from the nose across both wings. The solid aircraft rapidly vanishes without slowing down or changing direction.
[6–8 seconds] The camera continues tracking the same trajectory. The invisible delta-wing shape remains clearly perceptible because the sky and clouds inside its exact volume ripple and bend like concentrated heat haze.
[8–10 seconds] The refractive aircraft streaks toward the distant horizon, leaving a narrow turbulent distortion wake before exiting the frame.
[CAMERA]
One continuous very wide lateral tracking shot. The aircraft stays relatively small within the expansive frame. Fast pan tracking, broad environmental scale, strong lateral movement, subtle handheld vibration, and realistic motion blur. No close-up and no chase-camera angle.
[LIGHT]
Bright high-altitude daylight, cool blue sky, white cloud field, natural atmospheric haze, restrained contrast, and realistic cinematic texture.
[EDIT]
One uninterrupted shot. No cuts, transitions, slow motion, or camera repositioning.
[AUDIO]
Audio: violent high-altitude wind rush, rapidly approaching jet roar, a powerful Doppler flyby, and a brief vacuum-like sound as the camouflage activates.
BGM: an original 10-second cue, 80% tension and 20% awe. Low sub-bass pulse and metallic air textures, building sharply during the flyby and fading into a low tone over the final 2 seconds. Do not imitate an existing melody.
[NEGATIVE]
Do not show an empty sky after cloaking. Maintain a coherent, fast-moving delta-wing refraction that bends the background inside the aircraft’s exact volume.
No glowing outline, glass aircraft, force-field bubble, pixel dissolve, smoke camouflage, teleportation, morphing, or reappearance.
No extra aircraft, weapons, explosions, contrails, subtitles, on-screen text, watermarks, platform logos, or stickers.
Do not introduce Chinese text, garbled characters, or misspellings.
FORMAT
15 seconds | 16:9 | photoreal live-action war thriller
Fictional Sahelian city at blue-hour dawn.
Urgent, human, suspenseful, non-graphic.
REFERENCE CONTROL
Image 1 = locked AMINA identity and wardrobe.
Preserve face, close-cropped hair, deep-brown skin, navy field jacket, silver earpiece, canvas satchel, analogue radio, boots and determined expression.
Image 2 = locked environment.
Preserve concrete bridge, dry riverbed, dense low-rise city, rooftop antennae, dusty atmosphere, cool-blue shadows and restrained amber dawn.
No real countries, flags, organizations or landmarks.
STORY
Amina is the last radio operator keeping an evacuation bridge open. Communications fail during a distant attack. She must activate the backup relay and transmit clearance before automatic controls close the bridge.
00–02s — CHANNEL OFFLINE
Extreme close-up inside a dark concrete communications room.
Amina grips the analogue radio as red emergency lights pulse. Dust trembles, cables quiver, broken static fills the room.
Automated voice:
“Emergency channel offline.”
She looks up and immediately moves.
02–04s — BRIDGE CLOSING
Wide aerial-oblique rooftop view.
Civilians cross the bridge through windblown dust.
Checkpoint signal flickers green → red as shutdown begins.
A distant shockwave moves through city haze. No visible casualties.
04–06.5s — RUN
Low handheld tracking behind Amina running through a damaged service corridor.
Emergency lights flash, dust falls, cables briefly spark, boots hit wet concrete.
She pushes through a heavy metal door into blowing dust.
06.5–09s — FLASHBACK
Two minutes earlier.
Medium side-profile on rooftop.
Amina sees the unstable bridge signal, manually rotates the antenna and connects her radio to backup power.
Slow camera push. Wind moves her satchel strap as first amber sunlight reaches her jacket.
09–12.5s — TRANSMISSION
Rapid cross-cut montage:
frequency dial locks →
antenna aligns →
bridge cables vibrate →
red signal flashes →
Amina speaks into radio:
“Channel Seven, keep the bridge open.”
Overhead shot: civilians continue crossing.
Sharp hard cuts, brief black frames, natural motion blur.
12.5–15s — SIGNAL RESTORED
Sound suddenly drops.
Amina listens on the rooftop.
Clear radio tone.
Checkpoint signal changes red → steady green.
Evacuation continues.
Amina exhales and watches the bridge as dawn reaches the city.
Slow push toward her face.
Cut to black.
CAMERA
Large-format live-action realism.
Precise nonlinear cross-cutting.
Handheld tracking, aerial-oblique coverage, restrained push-ins and close details.
Natural motion blur, fine film grain, subtle gate weave.
No impossible camera movement.
ENVIRONMENT + PHYSICS
Cool-blue shadows against muted amber sunrise.
Deep blacks, dusty atmosphere, weathered concrete.
Wind realistically affects dust, clothing, cables and antenna equipment.
All movement and interactions remain physically believable.
AUDIO
Stereo radio static, electrical hum, wind, boots, antenna creaks, distant low artillery and helicopter rotors.
Restrained suspenseful sub-bass builds through the montage.
VISUAL
Photoreal cinematic war-thriller footage.
Immense scale contrasted with one determined operator.
Grounded, tense and human.
No subtitles, text, logos, watermarks, political messaging, flags or military insignia.
CONTINUITY
Same Amina identity, wardrobe, radio and satchel throughout.
Same bridge, city, atmosphere and dawn lighting.
Maintain geographic and temporal continuity across all shots.
Claymation. A whale breaches from a stormy clay sea and rises in a slow-motion arc above the waves, body fully extended, water sheeting off its flanks. At the top of the arc it rolls onto one flank, hangs for a beat, then falls back toward the surface. Midair, the camera sweeps beneath its belly in a bold dynamic move, revealing the vast churning ocean far below and the full stretch of its clay body against the sky, then holds as it crashes down and white clay foam bursts outward across the frame. Visible fingerprints, tool marks, and material imperfections in the clay. Slightly stepped, handmade movement. The sea is sculpted clay with rolling modelled waves and white clay foam. Audio: a deep rush of water, wind, the dull slap of clay on clay, distant gulls. Composition: subject centered, action contained within the central third of the frame.
Use the supplied image as the exact opening frame. Create a hilarious, high-budget 1980s live-action family comedy movie scene, photographed on a real soundstage with practical robot costumes, animatronics, handmade props, vintage wardrobe, and authentic 35mm film color.
The entire group suddenly hears something above them. Everyone’s eyes dart upward at the same moment. The men, women, children, and robots dramatically tilt their heads back, point toward the ceiling, and erupt into chaotic celebration. They gasp, scream, laugh, jump, wave their arms, slap each other on the shoulders, and completely lose their minds with exaggerated 1980s comedy reactions. The children bounce excitedly. The adults stumble into one another. The robots flash their eyes, flap their mechanical arms, spin clumsily, and celebrate with funny practical animatronic movements.
One man shouts:
“OPEN SOURCE?!”
The entire crowd answers together:
“MINIMAX H3!”
They cheer wildly as one robot attempts a victory dance, loses its balance, and is caught by the shocked family at the final second.
Fast, energetic comic timing with believable ensemble choreography. Begin with a brief locked group shot, then a subtle handheld push-in as the chaos escalates. Keep every original person, robot, shirt design, and the large headline clearly recognizable. Preserve facial identity, wardrobe, composition, lighting, and 1980s production design.
Authentic optical softness, warm tungsten lighting, rich film grain, slight gate weave, practical effects, natural motion blur, expressive physical acting, polished studio-comedy sound mix, triumphant synthesizer sting, cheering, robot beeps, and comedic percussion.
No CGI, no digital-looking robots, no morphing, no extra people, no duplicated characters, no distorted faces, no rewritten text, no spelling changes, no modern clothing, no style shifts, and do not turn the scene into animation.
A 16:9 cinematic wuxia mystery set in a bamboo forest at night. Use a low-saturation palette of cold blue, ink green, charcoal, and gray. Thin mist fills the forest and fine snow drifts through the air. The mood is austere, lethal, and controlled, with the tension of a martial-arts sect investigating a case and exchanging secret intelligence.
Deep in the forest, dense vertical bamboo fills the background while cold white mist-light glows in the distance. Soft, out-of-focus leaves partially obscure the foreground, creating the sense of watching from within the grove. A cool, soft front-side key lights the actors’ faces; backlight keeps the foreground dark and the distance luminous. Use shallow depth of field so leaves, snow, and bamboo dissolve into soft bokeh.
Prioritize facial close-ups and measured shot/reverse-shot coverage. Keep the rhythm restrained but tense. Photoreal period-drama production value, cinematic lighting, and no modern elements. No subtitles, on-screen text, watermarks, modern clothing or architecture, animation styling, over-smoothed skin, bright daylight, or comic performance.
16:9, 15s, hyper-realistic Korean noir crime teaser.
A rain-soaked Korean woman (late 20s, trench coat) enters an abandoned underground nightclub searching for her missing sister. A scarred crime boss (40s), presumed dead for years, emerges from the shadows with a lit cigarette. Atmosphere: 90s Korean noir, practical neon, rain, smoke, film grain, tactile realism, no CGI gloss.
Shots:
1. She enters: "언니... 여기 있어?"
2. Lighter ignites, revealing his scar. "오랜만이네."
3. She freezes: "당신... 죽은 줄 알았는데."
4. Slow push-in. He smiles: "네 언니가 날 죽였다고 생각했겠지."
5. They lock eyes. Thunder. Cut to black.
Audio: Rain, thunder, jazz crackle, lighter click, intimate silence during dialogue.
Create a 15-second ultra-premium cinematic luxury perfume commercial designed to fully showcase MiniMax H3's multimodal capabilities.
Output Settings:
• Duration: 15 seconds
• Resolution: 2K (1440p)
• Frame Rate: 24 FPS
• Aspect Ratio: 9:16 Vertical
• Native Stereo Audio
• Commercial-grade quality
Scene 1 (0–3s)
Extreme macro shot of a crystal luxury perfume bottle resting on polished black marble. Golden sunrise light passes through the glass, creating realistic caustics and sparkling reflections. Tiny perfume particles float naturally in the air. Slow cinematic push-in with perfect depth of field.
Scene 2 (3–6s)
A luxury fashion model wearing an elegant black dress walks confidently through a futuristic golden hallway. The camera smoothly circles around her while she gracefully picks up the perfume bottle. Hair moves naturally, realistic fabric simulation, premium fashion campaign quality.
Scene 3 (6–10s)
She sprays the perfume. The mist transforms into flowing golden particles that evolve into blooming flowers, liquid gold ribbons, shimmering butterflies, floating crystal fragments, and elegant light trails surrounding her. Every movement is physically realistic with seamless cinematic transitions.
Scene 4 (10–13s)
Hero beauty sequence. Slow-motion close-ups of the model's face, glowing skin, elegant eyes, glossy lips, luxury jewelry, and the perfume bottle. Beautiful bokeh, premium studio lighting, realistic reflections, cinematic lens movement.
Scene 5 (13–15s)
Epic ending. The perfume bottle rotates slowly on a glossy black pedestal surrounded by floating golden particles and soft luxury lighting. The model stands elegantly in the background. Premium cinematic composition worthy of a global luxury brand campaign.
Visual Style:
Luxury brand commercial, Dior Beauty, Chanel, Tom Ford, Apple-level production quality, Hollywood cinematography, ultra-photorealistic, ray tracing, global illumination, volumetric lighting, realistic reflections, premium color grading, cinematic depth of field, physically accurate materials, ultra-detailed textures, smooth camera movement, flawless facial animation, natural body motion, production-ready advertising quality.
Negative Prompt:
Low quality, blurry, noise, flickering, unstable motion, distorted anatomy, bad hands, bad face, duplicate objects, watermark, logo, text overlay, compression artifacts, oversaturated colors, unrealistic physics, low-detail textures.
Camera: First-person perspective at eye level with authentic handheld player movement, as if recorded directly from a modern AAA military shooter. The player carries a highly detailed assault rifle with realistic animations, visible hands, tactical gloves, dynamic reload mechanics, and weapon sway. **Opening Action:** The video immediately begins with the player already aiming down a roadway inside a modern military base. Multiple enemy soldiers are visible in the distance near sandbags, barricades, and military vehicles. The player carefully tracks one target, making small aim corrections while maintaining ADS (aim down sights). Fire several controlled bursts immediately at the visible enemies, producing realistic muzzle flashes, shell casings ejecting, smoke, recoil, hit reactions, and dust impacts around the targets. Continue firing in multiple short bursts while adjusting aim between enemies, simulating authentic FPS gameplay rather than scripted animation. **Movement:** After the opening firefight, lower slightly from ADS and begin advancing cautiously along the road beside concrete barriers, Hesco walls, and parked military vehicles. Frequently check left and right corners, briefly stop to reacquire targets, then raise the weapon and fire additional controlled bursts whenever enemies appear ahead. Continue pushing forward with deliberate player-controlled movement, using cover naturally and maintaining believable tactical pacing. **Environment:** Large modern military base with guard towers, armored vehicles, shipping containers, blast barriers, damaged buildings, smoke plumes, burning debris, scattered shell casings, dust clouds, and atmospheric battlefield haze. Cool natural daylight mixed with smoke and orange firelight creates a cinematic battlefield atmosphere. **Camera Motion:** Authentic player-controlled movement with subtle head bob, weapon sway, natural mouse-look adjustments, small left-right corrections while aiming, realistic recoil, smooth tracking of moving targets, brief pauses before shooting, and fluid forward progression. Avoid cinematic camera moves—everything should feel like genuine live gameplay captured by a skilled player. **Visual Quality:** Ultra-photorealistic, AAA game graphics with realistic PBR materials, detailed weapon models, physically accurate lighting, volumetric smoke, dynamic particle effects, crisp textures, realistic bullet impacts, muzzle flash illumination, motion blur only during rapid movement, and high-end military shooter presentation. **Gameplay UI:** Display a realistic modern FPS HUD inspired by games like PUBG, Battlefield, or Call of Duty (without copying exact copyrighted assets). Include: * Central dynamic crosshair or reticle * Ammo counter with magazine and reserve ammunition * Fire mode indicator * Compass at the top * Squad/team status panel * Mini-map in the upper corner * Health bar * Tactical equipment icons (grenades, medkit) * Hit markers when bullets connect * Directional damage indicators * Kill notification feed * Objective marker in the distance * Subtle interaction prompts and realistic HUD animations The HUD should feel polished, modern, and fully integrated into the gameplay, enhancing the illusion of authentic recorded footage from a contemporary military FPS.
