Raindance AI Prompts (Copy & Paste)
These prompts work best with a motion transfer model and a performance video you own. A text prompt alone cannot reliably recreate a specific dance.
Solo: replace the complete performer
In @Video1, replace the main performer with the complete person from @Image1. Use the uploaded person's face, hair, body proportions and clothing. Preserve the source choreography, gestures, camera movement, framing, scene, lighting and timing. Keep the same shot without new cuts. Do not add text, logos, music or speech.
Duet: keep each person assigned to one role
In @Video1, replace the performer who starts on the left with the complete person from @Image1. Replace the performer who starts on the right with the complete person from @Image2. Keep these identities assigned to their original roles when they cross positions. Use each person's own appearance and clothing. Preserve the choreography, camera path, timing, background and lighting. No audio, text or logos.
Preserve the selected scene
Keep the source's wet pier, rain, dusk lighting, camera composition and color grade. Change the complete performer only. The source video defines the dance; do not invent a new performance.
These are our own draft prompts for a video-editing model. They have not been visually tested with the new provider. Only use reference footage and photos you have the rights to use.