Short videoTool: MiniMax
Still poster → motion graphics in three directives (MiniMax H3)
A two-stage recipe: a strong still (here from @mmmiyama_D's poster prompt) + three plain-language motion directives, expanded to a full H3 prompt by an LLM. The insight: when the image is good, animating it is enough — don't redesign. Works for any typographic poster with a character.
OUR REPRODUCTION
Reproduction pending
SOURCE
@ponzponz15
PROMPT
Turn the still poster {poster_image} into a short vertical motion-graphics clip for MiniMax H3. Three directives only: 1. Horizontal scroll — the camera travels sideways across the poster's route-map graphics and lettering. 2. The character reads the world — the person recognizes the type and graphic elements around them and reacts: dodges a word, glances at a symbol, keeps walking through the layout. 3. Ending — a "foreground bokeh" style transition: a blurred shape sweeps across the frame and resolves into the logo {logo}. Expand these three directives into a full shot-by-shot H3 prompt (timecodes, camera, character action, type behavior, sound) with an LLM tuned for H3 before submitting. Keep the poster's palette, typography and layout exactly; animate, do not redesign.
VARIABLES
- {poster_image} · image · e.g. a typographic fashion poster (here: made with @mmmiyama_D's profile-screenshot → poster prompt)
- {logo} · text · e.g. MIYAMA
WORKS FOR
- Typographic posters with one character and graphic elements to react to (verified by the poster's output)
WEAK FOR
- Photographic stills without graphic elements — directive 2 has nothing to react to (hypothesis)
#motion-graphics#poster#minimax-h3#image-to-video#kinetic-type#two-stage