VIDEO2ROBOT
// End-to-end pipeline converting generative videos (Veo, Sora) to humanoid robot motions
Video2robot
End-to-end pipeline converting generative videos (Veo, Sora) to humanoid robot motions
13EmergingUnknown
What it does
End-to-end pipeline: Video (or Prompt) → Human Pose Extraction → Robot Motion Conversion Demo Video Backflip Dance Motion This project requires two conda environments: and . For details, see GMR README. For Blackwell GPU (sm120) users: For other GPUs (Ampere, Hopper, etc.): For details, see PromptHMR README. > Note: Scripts automatically switch to the appropriate conda environment ( or ) as
Getting Started
git
git clone https://github.com/AIM-Intelligence/video2robot