Fg-T2M: Fine-Grained Text-Driven Human Motion Generation via Diffusion Model.
Yin WangZhiying LengFrederick W. B. LiShun-Cheng WuXiaohui LiangPublished in: ICCV (2023)
Keyphrases
- fine grained
- human motion
- diffusion model
- coarse grained
- human body
- motion capture
- optical motion capture
- video sequences
- information diffusion
- anisotropic diffusion
- access control
- spatio temporal
- diffusion process
- human actions
- motion recognition
- image sequences
- keywords
- text mining
- action recognition
- diffusion tensor
- motion capture data
- visual data
- social networks
- pairwise
- machine learning