Co-Speech Gesture Video Generation via Motion-Decoupled Diffusion Model.
Xu HeQiaochu HuangZhensong ZhangZhiwei LinZhiyong WuSicheng YangMinglei LiZhiyi ChenSongcen XuXiaofei WuPublished in: CoRR (2024)
Keyphrases
- diffusion model
- space time
- video sequences
- key frames
- video data
- anisotropic diffusion
- multimodal interfaces
- motion history images
- information diffusion
- visual data
- image sequences
- hand motion
- diffusion models
- hand gestures
- motion analysis
- motion estimation
- optical flow
- video content
- gesture recognition
- laplace transform
- audio visual
- video frames
- human motion
- diffusion process
- motion trajectories
- video streams
- motion model
- human computer interaction
- human actions
- motion field
- motion segmentation
- moving objects
- three dimensional
- motion capture
- partial differential equations