mozgi ← All prompts
Seedance 2.0

A 15-second video built from a 3×3 nine-panel reference grid, each panel expanded to full screen while keeping the character and style consistent

Prompt

Reference @image1 Shared 3×3, 9-panel reference image
[CONDITION DEFINITION]
Use the attached 1:1 square reference image with 3 columns × 3 rows and 9 panels as a storyboard for a video of approximately 15 seconds. Do not animate the entire reference image as a 3×3 split screen; instead, present each panel individually as a normal full-screen shot. Use all 9 panels in the reference image without omitting, combining, repeating, or rearranging any of them. Show only one panel full-screen at a time; never show multiple panels simultaneously. If the reference image is live-action, preserve its live-action texture. If it is anime or an illustration, preserve the original style, lines, coloring, shading, and texture. Do not convert live-action to anime or anime or illustration to live-action. Do not introduce specific new clothing, expressions, poses, actions, props, or background settings in the prompt. Add natural movement suited to the content and style of each panel, guided by its composition, subject state, posture, expression, clothing, background, and atmosphere. If the 9 panels depict the same person or character, maintain their identity throughout all shots, including facial features, hairstyle, apparent age, build, clothing, and key visual traits.

[SHOT / FLOW]
Read the reference image from left to right, starting with the top row, then proceeding to the middle and bottom rows.
Shot 1|0.0–1.5 seconds
Present the top-left panel as a normal full-screen shot, adding natural movement in accordance with that panel’s composition, style, and subject state.
Shot 2|1.5–3.0 seconds
Present the top-center panel as a normal full-screen shot, adding natural movement in accordance with that panel’s composition, style, and subject state.
Shot 3|3.0–4.5 seconds
Present the top-right panel as a normal full-screen shot, adding natural movement in accordance with that panel’s composition, style, and subject state.
Shot 4|4.5–6.0 seconds
Present the middle-left panel as a normal full-screen shot, adding natural movement in accordance with that panel’s composition, style, and subject state.
Shot 5|6.0–8.0 seconds
Present the center panel as a normal full-screen shot, adding natural movement in accordance with that panel’s composition, style, and subject state.
Shot 6|8.0–9.5 seconds
Present the middle-right panel as a normal full-screen shot, adding natural movement in accordance with that panel’s composition, style, and subject state.
Shot 7|9.5–11.0 seconds
Present the bottom-left panel as a normal full-screen shot, adding natural movement in accordance with that panel’s composition, style, and subject state.
Shot 8|11.0–12.5 seconds
Present the bottom-center panel as a normal full-screen shot, adding natural movement in accordance with that panel’s composition, style, and subject state.
Shot 9|12.5–15.0 seconds
Present the bottom-right panel as a normal full-screen shot, adding natural movement in accordance with that panel’s composition, style, and subject state. End with a brief lingering moment.
Each shot should show natural movement from the very beginning, however subtle. The subject’s gaze, expression, breathing, posture, hair, clothing, surrounding objects, light, and background should move naturally from their original state in each panel.
Do not add new acting or large movements unrelated to the reference image.

[CAMERA / EDITING]
Generally preserve each panel’s original composition and sense of distance, naturally expanding it into a normal 1:1 full-screen shot.
Prioritize natural movement in the subject and background, and keep the camera stable by default. Only when necessary, add a very subtle push-in, pull-back, pan, or change in depth. Follow the timeline specified in [SHOT / FLOW] for shot order and transition times.
Switch between shots with clear cuts. Do not use morphing that dissolves the face, body, clothing, or background into the next shot. Do not retain the borders or margins separating the 3×3 image as a fixed frame in the finished video.
Decorations, lines, symbols, and effects already present in each panel may be naturally preserved as part of the original style. However, do not multiply new decorations, distort their shapes, or carry them continuously into another shot.

[SOUND]
With BGM. Add instrumental music that naturally suits the overall style, atmosphere, and pace of the video.
Do not include lyrics, singing, narration, or dialogue. If needed, add subtle ambient sounds or sound effects that do not interfere with the BGM.

[NEGATIVE]
Do not animate the entire 3×3 reference image as a split screen.
Do not use multiple screens, split screens, or a collage layout.
Do not omit, combine, repeat, or rearrange any of the 9 panels.
Do not change the specified reading order.
Do not deviate significantly from the specified timing.
Do not leave each panel static for a long time.
Do not change the identity of the same person or character.
Do not add new people, clothing, props, backgrounds, or acting unrelated to the reference content.
Do not significantly change the original expression, posture, clothing, or composition.
Do not turn live-action into anime, or anime or illustrations into live-action.
Do not dissolve or morph faces, bodies, clothing, or backgrounds between shots.
Do not multiply or distort decorations, lines, symbols, or effects, or connect them to another shot.
Avoid intense camera movement, extreme zooms, overacting, and unnecessary scene transitions.
  • Needs your photo: attach an image to the request — the prompt works with it.
Open in catalog
The rights to this prompt belong to its author @ai_lifehack55. You are responsible for any commercial use — promoting AI tools, aggregators or Telegram bots: mozgi.io accepts no liability. Details →
Model
Seedance 2.0
Format
1/1
Author
@ai_lifehack55
Original
x.com ↗

Similar prompts

All prompts for Seedance 2.0

Copied