CHARACTER LOCK — HIGHEST PRIORITY: The person in every shot is exactly the person in @image1. Preserve their facial identity, facial structure, eye shape and spacing, nose, mouth, exact skin tone, hairline, hairstyle, hair length, texture and colour, freckles and distinctive marks, apparent age, build and body proportions — identically, shot to shot. Preserve their gender presentation exactly as shown; do not masculinize, feminize or reinterpret it. Do not beautify, slim, de-age or restyle them. The attached image outranks every word below it: wherever the image and the text disagree, follow the image. Frame 1 is already in the scene — never a studio portrait, grey backdrop, or the attached still as an opening plate. Use they/them throughout. No other people. An exactly 15-second ultra-realistic cinematic dance performance film, 16:9. Hard cuts. Photoreal music-video. @image1 is the lead. Preserve her face exactly. High tight bun, long dark hair. Do not recast. @image2 is her wardrobe. Structured white strapless fashion top, dark navy high-waisted baggy jeans, black boots, chunky silver hoops. Same look every shot. Fully clothed. SET: underground parking garage. Board-formed concrete, low ceiling, fluorescent strips, wet polished floor. Two muscle coupes left and right as gateposts. One car at the back, headlights ON into the lens. LIGHT: backlit. Headlights plus two hard beams through haze, a V converging on the lead. Rim-lit, cool concrete, warm tungsten. Wet floor mirrors beams. CREW: eight dancers, mixed, exact unison, black leather jackets, black tops, wide light-wash jeans, combat boots. Left mirrors right. Faces not featured. Every wide shot centred, symmetrical, one-point. Vanishing point behind the lead. AUDIO: no music, no dialogue. Boots on wet concrete, fabric, breathing, car idle, garage echo. [0.0-2.5s] Wide, eye level, locked. Lead far back centre, small, rim-lit. Eight dancers a tight wedge, bent low, heads down, still. Beams a V at her. Only haze moves. [2.5-5.0s] Camera pulls back and lowers. Lead walks forward. Dancers break into two mirrored groups, low and heavy, arms sweeping. [5.0-6.5s] Low close. Lead large centre, walking toward lens, hands on hips. Two dancers at the left and right edges, arms out, framing her. [6.5-7.5s] HARD CUT. Beauty close-up, three-quarter profile, long lens, background dissolved. Hoop earring catches light. Jaw set. She does not look at camera. [7.5-8.3s] HARD CUT. Low wide. Lead centre mid-move, crew braced, unison. [8.3-9.5s] HARD CUT. Close-up of one black combat boot on wet concrete, laces sharp, rim light. [9.5-13.0s] HARD CUT. Extreme low angle on the floor, tilted up. Lead in a low dance move, crew around her, arms driving, mirrored. Wet floor reflections. [13.0-15.0s] HARD CUT back to first framing. Lead centre, both arms raised in a V. Eight dancers bowed at the waist in two mirrored lines. Starburst headlights. Hold. NEGATIVE: on-screen text, logos, brand names, music, extra limbs, face drift, changing clothes, looking into the lens, rain, fire, broken symmetry.
Drag the clip sideways
CHARACTER LOCK — HIGHEST PRIORITY: The person in every shot is exactly the person in @image1. Preserve their facial identity, facial structure, eye shape and spacing, nose, mouth, exact skin tone, hairline, hairstyle, hair length, texture and colour, freckles and distinctive marks, apparent age, build and body proportions — identically, shot to shot. Preserve their gender presentation exactly as shown; do not masculinize, feminize or reinterpret it. Do not beautify, slim, de-age or restyle them. The attached image outranks every word below it: wherever the image and the text disagree, follow the image. Frame 1 is already in the scene — never a studio portrait, grey backdrop, or the attached still as an opening plate. Use they/them throughout. No other people. An 8-second ultra-vertical 9:16 ADHD-cut fashion film. Super-fast cuts, massive sound design. Photoreal + light-streak graphic. @image1 is the FIRST FRAME: extreme close-up of THIS woman's real eye. Keep that exact iris colour, lashes, skin. Do not turn it green. Do not recast. @image2 is the same woman in the burgundy faux-fur coat. Preserve her face. She WEARS the coat in every face/body flash. NO LOGO. No emblem, no letters, no Nike, no swoosh. No green eye. No cat eye. No animal eye. AUDIO: huge hits, whooshes, sub drops, no lyrics, no voiceover. [0-1.5s] HOLD the first frame then a tiny live micro-move: HER eye, real colour, looking into camera. Smash. [1.5-4s] Chrome light-streaks, smoke, then HARD CUT — her full face fills the frame, powerful, burgundy fur collar. Smash. [4-6s] Coat texture macro, burgundy pile, gold bangle, then her three-quarter in the coat, ripped away by light. [6-8s] Faster: face / coat / black. Final hold: her face in the coat, or black. NEVER a logo card. NEVER a green eye. NEGATIVE: green iris, cat eye, animal eye, any logo, extra people, Nike, slow holds.
Two figures appear throughout, each with their own distinct outfit — one built on a black graphic-collage look, one on a white brushstroke look. Hold each of them completely consistent with themselves across every shot: same face, hair, body, garments, prints, colours, hardware and footwear, with no drift or reinterpretation between shots. Every shot is a graphic fashion poster: the full-length figure or figures standing on a flat, hard-edged colour-block ground with heavy condensed all-caps display type set large beside or behind them, in one consistent palette of off-white, vermilion red, burnt orange, deep teal, dense black and cool grey. Polished stylized 3D character animation throughout — smooth matte skin with no visible pores, refined idealized features, hair as clean sculptural masses, garments keeping real fabric behaviour but surfaced smooth rather than photographic: hard crisp studio light, grounded contact shadows, true material separation, consistent grade across every shot. The graphic ground and the display type in each shot stay flat, sharp and stable — they are printed backdrop and set lettering, never overlaid captions, never animated titles, never a band or bar across the frame. Motion is restrained and continuous. Each shot holds one figure action and one slow camera action, evenly weighted, letting the pose read. Shot 1 — the poster reads SO CLOSE. the lone figure settles its weight from one hip to the other, loose fabric swinging and falling back to rest. The camera pushes in slowly. Shot 2 — the poster reads PATHS CROSS. the two figures walk past one another in opposite directions, neither turning to look, loose fabric trailing behind them. The camera tracks sideways against their direction. Shot 3 — the poster reads FARTHER NOW. the lone figure draws one hand from its pocket and lets it drop, shoulders easing down. The camera drifts slowly to the side. Shot 4 — the poster reads STILL APART. the two figures, one on each side of the dividing line, each shift a half step further from the divide, turning fractionally away. The camera pulls back slowly. Shot 5 — the poster reads OUT OF SYNC. the lone figure lowers a raised arm across the body, the sleeve falling open, the words mounted as large dimensional letters on the wall behind them holding still. The camera pushes in slowly past the figure. Shot 6 — the poster reads ONE BEAT OFF. the two figures step apart in opposite directions, one beat out of time with each other. The camera eases upward. Shot 7 — the poster reads PASSING BY. the two figures stand back to back and each turn their head slightly away, smoke ribbons drifting across behind them. The camera drifts slowly upward. Shot 8 — the poster reads ALMOST US. the two figures each lean a fraction toward the other and stop short, the translucent ribbons crossing and parting between them. The camera pushes in slowly on the gap. Shot 9 — the poster reads BETWEEN. the two figures turn to face one another and hold, the word standing still between them. The camera pulls back slowly to leave both in frame. Hold every figure fully in frame from head to toe in every shot, feet inside the frame. Keep the two figures distinct from each other and consistent with their own reference across the whole piece. Do not invent any words, lettering, numbers, logo, signature or watermark beyond the nine poster lines named above. No subtitles, no captions, no title cards, no lower thirds. No extra people. No photographic skin texture or visible pores, no cel shading, no ink outlines, no comic or manga line-art, no flat vector fills, no painterly or sketch rendering. No handheld shake, no whip pans, no zoom snaps, no speed ramps, no strobing, no lens flare.
A winter coat closet, rendered as a stylized animated feature. A tiny steam locomotive bursts out of a hanging coat's pocket in a spray of snow, a mouse conductor leans from the cab and calls the last train, and pocket seventeen opens onto a snowy tunnel glowing blue. Follow the train through and you land on the platform of a miniature railway station hidden inside the coat's quilted lining, where a mouse child is waiting with one lost mitten and no match for it. Pick or create your avatar, build a turnaround sheet so the character holds through every shot, lay the twelve beats out as a storyboard, then render the finished film.
15-second 9:16 photoreal. @image1 is FACE ONLY — photocopy their face. Do not use a turtleneck. Outfit: grey hoodie, beige shorts, skateboard. @image2 is the landscape — photocopy that cliff road AND the completely fantasy skyline (floating palaces, impossible spires, aurora, turquoise bay). Not a real-world city. 0–3s: Close on the face, wind, they drop in on the skateboard at the top of the cliff road. Fantasy skyline already visible behind. 3–8s: High tracking down the steep switchback. Speed, dust, board on asphalt toward the water. Unreal towers fill the horizon. 8–12s: Wide: tiny figure racing the cliff edge, bay and fantasy skyline below. 12–15s: They ride toward the shore. Hold the fantasy skyline. No on-screen text. AUDIO: wheels, wind, no lyrics.
@image1 is ELARA — photocopy her exact face, silver-grey wavy hair, brown eyes, AND this exact outfit: black leather bomber (open), white tank, ripped denim shorts, black over-the-knee stiletto boots, sunglasses on her head, silver chain. Do not put her in a white tube top. Do not give her arm tattoos. Do not bun her hair unless a brief beat. She MOVES — hip-hop dance, not a frozen JPEG of the still. @image2 is the LOCATION — photocopy this sunset graffiti alley: wet asphalt, golden sun at the vanishing point, tall brick buildings, colorful graffiti on both walls, puddles, long shadows. She dances IN this alley. Not a pink flower studio. Not neon indoor plants. ONE PERSON. No extra dancers, no statues, no second body. 30-second 16:9 music-video energy: dope original hip-hop (no named commercial song), multi-cut, handheld + smooth orbits, beast-mode intensity. No on-screen titles. No end card. [TIMELINE — 30 SECONDS] 0–5s: Wide 16:9. ELARA in the @image1 outfit stands in the @image2 alley, sun flaring behind her. She hits the first groove — shoulder, step, hair moves. 5–12s: Medium. Hip-hop: body rolls, sharp hits, walk-and-turn down the wet street. Bomber and boots read. Graffiti walls in frame. 12–18s: Low angle, sun in lens. She drops into a wide stance, arms, then pops up. Puddle reflections. 18–24s: Tighter cuts: boot stomp, bomber sleeve, silver hair whip, face (ELARA, not another woman). 24–30s: Pull wide. She hits a final pose walking toward camera, sunset alley full-frame. Hold one beat. Cut. Audio: original dark hip-hop beat + alley ambience. No lyrics from a famous song. No voiceover.
30-second 16:9 ultra-realistic AAA third-person action gameplay. Sharp Unreal-like capture, not a cinematic montage. @image1 is absolute character truth (face and sci-fi suit). @image2 is the environment. Photocopy both. Do not describe appearance. Use they / their / them only. SCI-FI world, not fantasy Korea. One colossal chrome-white mech tiger guardian. 0–4s: Sprint up damaged metal stairs toward the fortress. Jump a broken span, grab a ledge, climb into a ruined courtyard. 4–8s: They pick up a glowing pale-blue relic in a shrine-like console. Structures and glyphs light. 8–12s: Ground cracks. A colossal mech-tiger guardian rises. Scale read. 12–17s: Guardian wrecks the gantry. They dodge falling beams, slide, jump a broken energy bridge, swing a cable over the ravine. 17–21s: Relic dash — fast blue trail across broken platforms. 21–25s: Guardian chase. Dodge-roll a energy strike. 25–28s: Short blade, strike the glowing weak point on the shoulder. Blue pulse. 28–30s: Guardian roars, they take a defensive third-person stance. Boss fight about to start. No on-screen text, no HUD, no logos. AUDIO: boots, metal, relic hum, roar, no lyrics.
@image1 is LANA — photocopy her exact face, bone structure, long straight blonde hair, AND this exact outfit: black satin strapless bandeau, matching black satin wide-leg pants, black leather bomber jacket slipping off one shoulder, large silver hoop earrings, layered silver necklaces, silver rings, black sneakers. She MOVES. Never a frozen JPEG. FORBIDDEN from @image1: OPENCLIPS.AI watermark, any on-screen logo. FORBIDDEN from the source recipe: pastel-pink hair, green pearl baseball cap, green cropped sweater, plaid shirt at the waist, baggy light-wash jeans. Do not recast as East-Asian pink hair. ONE PERSON. Empty fields and playgrounds. No extra faces. Cinematic 30-second 16:9, photorealistic, melancholic and dreamy, muted greens, concrete greys, late-afternoon light. Slow pans, gentle dollies. No on-screen titles. No lyrics. No karaoke. No Indonesian text. [TIMELINE — 30 SECONDS] 0–3s: Medium-wide. LANA in the @image1 satin/bomber look lies on her side on weathered grey stadium steps, head on the riser, looking softly at camera. Empty sandy field. 3–6s: She hangs from the white soccer-goal crossbar, both hands on the bar, body dangling, looking at camera. 6–9s: Inside a stainless-steel cage packed with orange basketballs, eyes closed, cheek on a basketball as if sleeping. 9–12s: Reclines against a red-yellow-green playground tube slide, eyes half-closed. 12–16s: Sits still on a chain swing, looking down, empty swing beside her. Blue rubber floor. 16–19s: Sits on a low concrete wall, orange traffic cone on her head like a hat, blank stare. 19–23s: Outdoor ping-pong. Holds a red-and-black paddle, then rests head and arms on the blue table, looking sideways. 23–26s: Stainless-steel drinking fountain. Presses the button, water arcs; she leans in then looks at the basin. 26–30s: Stadium steps. Walks past a black backpack, looks at the sky, sits and leans back, eyes closed, legs extended. Hold. Audio: quiet outdoor ambience — wind, distant playground, no song, no vocals, no Indonesian lyrics.
@image1 is MATEO — photocopy his exact face, bone structure, wavy chestnut hair, stubble, AND this exact outfit: black tuxedo jacket, open ivory dress shirt, loose black necktie, black trousers with satin side stripe, patent black oxfords. He MOVES. Never a frozen JPEG of the still. SET: elegant luxury hotel — dark marble, gold railings, chandelier bokeh, high-end night interior. Not a pale blue studio. @image2 is the BOTTLE. Photocopy this perfume exactly: matte black rounded-rectangular body, matching matte-black cylindrical cap, recessed black label plate with white type TOM FORD / OMBRÉ LEATHER as in the photo. Same silhouette. NEVER a Bvlgari rectangle with a rose-gold cap. NEVER a navy Chanel cube. ONE PERSON. No extra faces. Cinematic luxury perfume commercial, 10 seconds, 16:9. Slow elegant movements, shallow depth of field, premium film. VOICEOVER (spoken, exact): “Getting ready, one final touch. Tom Ford Ombré Leather, make it yours.” No on-screen titles. No Chanel. No Bvlgari. No end-card logo. [TIMELINE — 10 SECONDS] 0–2s: MATEO in the @image1 tux, luxury marble behind him, holds the @image2 matte-black Tom Ford bottle with both hands toward camera. 2–4s: Close-up of his fingers lifting the matte-black cap, spray nozzle. 4–6s: He sprays a fine mist on his inner wrist. Bottle identity stays @image2. 6–8s: He lifts the wrist to his face, eyes close, small satisfied inhale. 8–10s: Bottle fills the frame, sharp; he is softly blurred behind it. Hold. Audio: the voiceover above + quiet room. No music bed that drowns the line.
@image1 is LANA — photocopy her exact face, bone structure, mouth, nose, brows. IGNORE ivory coat and sunglasses in @image1. WARDROBE (from the brief, not the source clip, not @image1): Silky white nightgown, hair wrapped in a white towel turban. Face visible. Luxury bedroom, high-end vanity desk, warm lamps, marble or lacquered wood, glass, gold hardware, soft morning light. NOT a living-room sofa. NOT a blue t-shirt. NOT a plant-and-couch set. @image2 is the PRODUCT. Photocopy this bottle exactly: taupe/beige fluted cylindrical body, matching pump, peach-gold label type as in the photo. Same silhouette, same pump, same colour. NEVER a white squeeze tube. NEVER a blue round logo. NEVER Nivea. The bottle is large and readable whenever it is on camera. She MOVES. Applies lotion. Never a frozen JPEG. Never open on a studio headshot. ONE PERSON. No on-screen titles. No end card. No overlay slogans. [TIMELINE — 30 SECONDS, 9:16] 0–5s: Close-up, vanity. LANA in white silk nightgown, towel on hair, looks down, rubs a pearl of lotion along her forearm. Soft luxury bedroom bokeh, lamp glow. 5–10s: Cut to hands over the vanity. She holds the taupe fluted pump bottle (@image2) and presses the pump into her palm. Cream-white lotion. Bottle identity must match @image2. 10–15s: She sets the bottle on the vanity (still visible, still @image2). Rubs palms together, then presses fingertips to her lips/cheeks softly — night-routine, not a commercial smile-to-camera yet. 15–21s: Medium. She sits at the vanity, lifts the taupe bottle toward camera, nightgown + towel, bottle large in frame, label readable as @image2. Warm smile. 21–26s: She works lotion into her arm, skin dewy. Vanity and bottle in the background. 26–30s: She raises the glowing arm toward camera, small smile, towel hair, white silk. Hold. No logo card. Audio: quiet room, pump click, skin. No music. No voiceover.
Night GelRecreate workflow