Back to MiniMax H3 prompts
Action reference / Reference Generation
Official manual translationTranslated from a MiniMax H3 manual exampleStreet-dance motion transfer
Let the video own choreography and the images own identity.
Actual H3 output / 10s
Prompt -> result
What this output demonstrates
The output transfers a fast street-dance routine to the character defined by the image references.
- Control pattern
- Let the video own choreography and the images own identity.
- Check in your run
- Using all references for all attributes can blend faces, clothes, and movement sources.
Shot structure
0s5s10s
INPUTR2V
ACTION01
CAMERA01
AUDIOON
Working principle
Let the video own choreography and the images own identity.
Watch for
Using all references for all attributes can blend faces, clothes, and movement sources.
Adapt this brief
Prompt draft
Use Video 1 strictly for the street-dance movement and timing. Use Images 1 and 2 for the performer identity, appearance, and wardrobe. Recreate the complete dance performance with the referenced person while preserving body rhythm and camera framing.Replace the subject and reference roles first. Change camera, timing, and sound one variable at a time.