Platinum-haired model walking by concrete wall AI Image Prompt
I. First-frame lock (Image-to-Video baseline) Use the reference image as the sole first frame, strictly keep the original composition, camera position, color tone, clothing, and facial appearance unchanged: Asian female model, extra-long platinum-blonde straight hair blown apart by wind, strands sweeping across her face, wearing an off-white knit beanie, true-red lip makeup, in an oversized emerald-green chunky knit sweater, dark brown wide-leg trousers, crossbody bright orange nylon soft bag; she passes in profile with a long-striding walking pose, head turned to camera, gaze direct; background is a gray fair-faced concrete wall (with regular holes), with a hard diagonal sunlight boundary shadow on the wall; in the left rear, a male passerby in a black down jacket and black backpack passes by blurred. Low-angle wide-angle upward view. II. Overall style anchor (unchanged throughout) High-fashion editorial / street documentary cinematic / stream-of-consciousness slow rhythm; color is high-contrast clashing palette—emerald green × bright orange × brown × off-white × cement gray; hard midday natural sunlight, sharp light-shadow division, no artificial lighting or light effects; 35mm film texture, real grain, shallow depth of field, slight handheld breathing, low-angle wide-angle slight distortion; camera movement restrained, slow, weighty; no magic/light effects/particles/lens flares/text subtitles/dialogue or lip sync throughout. III. Storyboard script (8 shots, continuous story) Shot 1 | 0.0–1.2s | Start (first-frame micro-motion) From the first frame, camera remains low-angle and moves extremely slowly dolly left; model maintains frozen walking pose with micro-motion: platinum long hair continues to flare backward in wind, sweater cuffs and trouser hems tremble slightly, orange crossbody bag sways slightly with body weight, gaze always locked to camera; background passerby continues walking blurred. Shot 2 | 1.2–2.4s | Footsteps Cut to low-angle leg close-up, brown wide-leg trousers roll open with long strides revealing fabric folds, soles press over concrete ground, the diagonal shadow boundary on the wall moves laterally behind her. Shot 3 | 2.4–3.6s | Profile Slow cut to side close-up, capture strands brushing lips and jawline, knit texture at beanie edge clear, red lips and cool gray wall form strong contrast; her gaze slowly moves from camera to forward. Shot 4 | 3.6–5.0s | Parallel tracking Cut to lateral tracking, moving along concrete wall at same speed as model, maintain full-body medium shot, wall holes and light-shadow boundary continuously sweep backward; background male passerby exits frame from left side. Shot 5 | 5.0–6.2s | Detail Cut to close-up of orange bag and hand, bag body sways slightly back and forth with stride, bag strap presses knit deformation into sweater, her left fingertips emerge from cuff and gently brush aside hair blown across chest. Shot 6 | 6.2–7.5s | Looking back Cut back to low-angle wide-angle medium shot, model slows pace, body turns sideways, head slowly turns back to look at camera again, long hair goes from flying to slowly settling, shoulder line relaxes, breathing rise and fall visible. Shot 7 | 7.5–8.7s | Pausing Camera slowly tilts up from knees to chest/shoulder position, model stands still in place, weight on one leg, wind continues to move hair tips and sweater fuzz, gaze calmly direct to camera. Shot 8 | 8.7–10.0s | Approach (ending) Camera continues slow push-in to facial and shoulder medium close-up, hair and light-shadow relationship remain unchanged, image steadily holds during push-in; ending maintains forward push momentum, no pullback, no zoom out, no rise, no fade to black. IV. Consistency and technical requirements Throughout, facial appearance, hair color and length, beanie, green sweater, orange bag, brown trousers remain completely consistent; no outfit changes, no face changes, no makeup changes. Background wall material, hole arrangement, diagonal shadow direction unified throughout; passerby always blurred, not clear, no interaction. Wind direction unified from front-right of frame, hair movement direction consistent throughout. Movements slow and weighty: walking → slowing → looking back → standing still, rhythm progressive without jumps. Transitions only with slow camera movement or natural slow cuts, camera position span no more than 40°, no jump cuts, flash cuts, deformation/clipping. 24fps cinematic, 9:16 vertical, 4K detail, realistic skin texture and knit fiber texture. Sound effects (optional, no voice): city ambience low frequency, footsteps, wind, fabric friction. V. Negative Prompt Light effects, flare spots, lens flares, glowing particles, dynamic effects, slow-shutter motion trails, text subtitles watermarks logos, dialogue and lip sync, exaggerated smile changes, outfit/hair color changes, facial distortion, facial feature drift, extra fingers, extra people, passerby clear and stealing focus, AI plastic skin, over-smoothing, overexposure high saturation, fake HDR, rapid shaking, rotating camera, ending pullback/zoom out/rise, frame flicker, style mutation.
Content: Structure = image-to-video first-frame lock using a reference image; style anchor (editorial street documentary, clashing colors, hard light); 8-shot storyboard with timing, camera moves, shot sizes; consistency rules; negative prompt.
Pros: Clear per-second shots with executable camera moves; detailed consistency and negative constraints.
Cons: Too dense for one generation; strongly tied to the reference image, limiting variation.
Reference image: Clearly relies on reference image as sole first frame.
Curated from publicly available creative examples. Prompts and commentary are edited and organized by this site. Please contact us with any rights concerns. Copyright & Privacy