Subjects: @image_1 is the first frame of this video and the only reference. The two performers in @image_1 are the whole band - the one holding the pastel-pink guitar is the guitarist, the one holding the baby-blue bass is the bassist. Identity: both performers in every shot are exactly the two people in @image_1 - faces, skin tone and gender presentation unchanged, never masculinized or feminized, never merged, swapped or duplicated, and the guitarist always keeps the guitar while the bassist always keeps the bass. The hair of each keeps exactly the length, texture and colour shown in @image_1: never lengthened, never restyled, and long hair is never added to a short haircut. Wardrobe and set: the matching pink school-uniform sets, the accessories, the instruments, the pastel practice room, the pink amplifier and the whole colour palette come from @image_1 and stay unchanged for the whole video. Style: high-key candy-coloured pop promo - soft bloom, gentle film halation, glitter specks in the light beams, saturated pastels with no muddy shadows. Vertical framing, one camera move per shot, hard cuts and no dissolves. Shot 1: The video starts exactly on @image_1 and the camera pushes in. Both performers hit the first chord on the downbeat and the tulle curtain lifts behind. <a big clean electric chord, the amplifier humming> （bubblegum pop at about one hundred forty beats per minute lands on the downbeat） Shot 2: A close shot travels up the neck of the pink guitar to the face of the guitarist, a fast bright riff, then a grin straight down the lens. <fast picking on wound strings, a pick scraping once> （the guitar line carries the melody） Shot 3: A low tracking shot passes the bassist, who leans into the microphone for a wordless harmony and turns once on the spot as the first pale pink confetti falls. <a round warm bass line, an airy wordless harmony, confetti fluttering> （drums come in punchy underneath） Shot 4: The camera arcs around both performers, back to back at the centre of the room with heads tipped back, confetti settling in the hair of each. <guitar and bass locked together, a shared laugh> （the chorus opens up wide） Shot 5: A rooftop at dusk under a cotton-candy sky. Both stand at the railing seen from behind with the cases on the ground and bump shoulders once as the city lights come up, the two figures small in the lower third and the sky filling the rest. <open wind, distant city hum, a last cymbal ring> （the song lands on its final beat and stops clean） Continuity: only the two people from @image_1 exist in this entire video and no additional figure is ever visible, not even in the distance. Same uniforms, accessories, hair length and instruments from the first shot to the last, five fingers on every hand. There are no lyrics and no words in any language anywhere in the audio - wordless harmonies only. Constraints: Keep it subtitle-free and avoid generating any text or subtitles anywhere. Do not generate a logo. Do not generate a watermark. Do not generate duplicate performers or a twin effect.

4 steps

subject_definitions: <Subject 1> is the product the reader adds in step 1, shown in @product. Its silhouette, its proportions, its colour, its finish and every mark printed or engraved on it come entirely from @product and not from this text. @product defines the product only; its background, framing and lighting are not carried over. summary: [reference generation] The target video is a fifteen-second vertical launch film in which <Subject 1> is shown on a seamless off-white paper backdrop across eight fast shots — a hero frame, three macro passes, two levitation frames, one demonstration and an end card — with four short lines of on-screen text written from the product itself. retention_analysis: <Subject 1> (appears in [Shot 1], [Shot 2], [Shot 3], [Shot 4], [Shot 5], [Shot 6], [Shot 7], [Shot 8]): fully_preserved - the silhouette, the proportions, the colour and the finish stay exactly as in @product in every shot, and its surface carries only the marks that exist there. detailed_description: The target video is a live-action, high-end product commercial in vertical 9:16, lit by soft daylight from the left against a seamless off-white paper backdrop, with shallow depth of field and a clean neutral grade. The objects on the paper are chosen by reading @product: two or three simple things that come from the same world as that product — the material it is made from, the thing it is used on, or the thing it is used with — and they are the only objects in the film besides <Subject 1>. They stay the same two or three objects from the first shot to the last. Four shots carry one line of on-screen text each, and each line is written by reading @product: two words in title case, at most sixteen characters, naming one quality that this particular product genuinely has. Each line sits in the vertically central zone in a bold sans-serif, the first word charcoal grey and the second word steel blue, rising into place over about fifteen pixels. One line is on screen at a time, and the other four shots carry only the image. [Shot 1] A wide frame holds <Subject 1> lying on the paper among those objects. The camera pushes in with large amplitude at fast speed until <Subject 1> fills the lower third. [Shot 2] At 00:02.000, the shot cuts to an extreme macro of the part of <Subject 1> that does its work — the edge, the tip, the opening, whichever part of this product meets what it is for. The camera trucks right with large amplitude at fast speed as a narrow specular highlight races along that part, and a line of text appears naming how well this product performs that job. [Shot 3] At 00:04.000, the camera cuts to <Subject 1> spinning in the air above the paper, those objects whipping past it. The camera holds a static shot while the product completes a full turn. [Shot 4] At 00:06.000, the shot cuts to a macro of the part of <Subject 1> a hand would take hold of. The camera trucks left with small amplitude at fast speed across that part, and a line of text appears naming how this product feels to hold. [Shot 5] At 00:08.000, the shot cuts to an extreme texture macro in which the broadest surface of <Subject 1> fills the frame. The camera pans right with large amplitude at fast speed as one band of light sweeps over it. [Shot 6] At 00:10.000, the shot cuts to a slow-motion side view of one demonstration: the simplest action that shows what <Subject 1> is for, performed on it with one of the objects from the paper. The camera holds a static shot. [Shot 7] At 00:12.000, the shot cuts to a levitation frame in which <Subject 1> and those objects hang at different heights above the paper and rotate together, each casting its own shadow on the paper below. The camera pulls out with small amplitude at fast speed, and a line of text appears naming what this product does for the person using it. [Shot 8] At 00:13.500, the shot cuts to the end card: <Subject 1> rests on the paper among those objects, nothing moving. The camera holds a static shot, and a line of text appears naming what this product is made of, and it remains to the final frame. overall_soundscape: A quiet room tone runs under the whole video at a low, even level. Each cut carries one short, soft sound made by the material of the product in @product itself, and each of those sounds sits just above the room tone. non_diegetic_music: A steady electronic pulse at about 128 beats per minute with a soft kick and clipped hats, held at one even, moderate level from the first frame to the last, ending on a single short tail.

2 steps510.3K

Style: luxury skincare commercial, minimal, clean, soft ivory and blush-pink palette, premium CGI, glossy liquid physics, elegant editorial beauty, photorealistic, slow motion. Product lock: a tall clear cylindrical pink-liquid toner bottle with a white screw cap and a white front label is the only hero. Match the reference photo exactly — bottle shape, proportions, cap, label, typography, colors, fill level, packaging. No redesign, no extra products, no on-screen text, no captions, no typography overlays. Liquid: pale translucent pink watery toner matching the bottle. Not milky, not white, not opaque. Eight beats in one continuous 10-second spot, 9:16. Beat 1. a tall clear cylindrical pink-liquid toner bottle with a white screw cap and a white front label slowly descends against a pristine ivory-white field with a faint warm pink gradient. Fine water droplets float around the bottle and catch the light. Camera: slow cinematic push-in. Lighting: soft diffused studio, delicate glossy reflections. Beat 2. The bottle settles into crystal-clear water. An elegant splash rises around the base in slow motion. Blush-pink caustics shimmer through the liquid. Camera: high-speed macro. Details: realistic water physics, fine droplets, glass-like refraction. Beat 3. The bottle rotates a full turn just above reflective water. Tiny droplets orbit it. The background eases from ivory into ultra-soft blush-pink. Camera: smooth product orbit. Details: sharp product focus, shallow depth of field, realistic reflections. Beat 4. Extreme macro of pale pink toner dispensing from the bottle. A silky translucent pink drop forms and falls. Camera: macro tracking the drop. Details: realistic viscosity, glossy surface tension, soft pink highlights. Beat 5. The drop lands on calm water. Concentric ripples, micro-bubbles, light sliding across the surface. Camera: slow-motion macro at water level. Mood: calm, fresh, hydrating. Beat 6. Extreme macro of the pink toner spreading across pristine glass. Silky, lightweight, luminous, with delicate waves and tiny air bubbles. Lighting: soft beauty-editorial, glossy highlights. Camera: slow macro track. Beat 7. One hand lifts a tall clear cylindrical pink-liquid toner bottle with a white screw cap and a white front label. Soft side sunlight, highlights along the bottle and cap. Camera: slow rotation around the hand and product. Background: minimal warm-white studio, faint blush-pink gradient. No other people. Beat 8. a tall clear cylindrical pink-liquid toner bottle with a white screw cap and a white front label stands centered on a glossy reflective water surface. Soft blush-pink mist, a few floating droplets, a premium finishing sparkle. Camera: very slow push-in toward the label. Lighting: soft diffused luxury studio, realistic reflections, cinematic depth of field. Final look: high-end skincare campaign, photorealistic product CGI, elegant, minimal, ivory-and-blush, realistic water and liquid physics, no distortion, no extra products, no altered packaging, no on-screen text.

2 steps685.9K

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. PRODUCT LOCK — HIGHEST PRIORITY AFTER THE PERSON: @image2 is the hero bag. It is the only bag. Preserve its exact camel-tan leather, bunny-ear knots, slim shoulder strap, gold hardware, gold ball charm, gathered silhouette and proportions. No other bag, no tote, no redesign. The bag is in every shot with the person — worn on the shoulder, held in the hand, or sitting against them on the floor. Never leave the bag out of frame. WARDROBE: keep the look from @image1 — ivory double-breasted blazer, white shirt, camel mini skirt, tall tan suede boots. Do not put them in a black blazer, shorts, white socks or loafers. SETTING: a clean luxury fashion studio, seamless cyclorama, warm ivory-beige walls, glossy pale floor with soft reflections, hard directional sunlight from one side. No crew, no photographer, no other people, no on-screen text, no captions. STYLE: high-fashion bag commercial, 9:16, 15 seconds. Copy this studio-fashion movement and camera language: full-body hero, overhead floor sit, over-shoulder look, walk with the bag, seated pose, squat, walk-away look-back. Slow editorial pacing, photoreal, shallow depth, commercial lighting. Frame 1 is already in the scene — never a studio portrait plate of the attached still. Beat 1 (0-2s). Full-body hero. They stand in the cyclorama, weight on one leg, the bunny-ear bag on the shoulder exactly as in the product photo. Slow push-in. Bag fully visible. Beat 2 (2-4s). Overhead. They sit on the glossy floor, knees folded, one hand on the bag beside them. Bag fills a clear part of the frame. Camera high above. Beat 3 (4-6s). Medium. They turn, look back over the shoulder at camera. Bag in the hand at their side, bunny ears readable. Beat 4 (6-8s). Product insert. Close-up of the bag in their hand: bunny-ear knot, leather grain, gold charm. They do not leave the shot — fingers and forearm stay in frame. Slow orbit. Beat 5 (8-10s). Walk. They walk across the cyclorama carrying the bag by the strap, a natural swing. Full-body, floor reflection. Bag never hidden by the body. Beat 6 (10-12s). Floor sit. They sit with one leg extended, the bag standing or held against the hip, facing camera. Plenty of negative space above. Beat 7 (12-13s). Low squat, facing camera, bag on the shoulder. Still, then a small weight shift. Beat 8 (13-15s). They walk away and look back over the shoulder, bag in hand. Hold on the look. Slow push-in. End. No extra products, no logos invented, no on-screen type.

3 steps391.8K