AttT2M: Text-Driven Human Motion Generation with Multi-Perspective Attention Mechanism.
Chongyang ZhongLei HuZihao ZhangShihong XiaPublished in: CoRR (2023)
Keyphrases
- human motion
- multi perspective
- attention mechanism
- human body
- image sequences
- motion capture
- optical motion capture
- video sequences
- motion recognition
- spatio temporal
- visual attention
- d scene
- human movements
- keywords
- motion capture data
- saliency map
- human actions
- motion synthesis
- humanoid robot
- visual attention model
- event detection
- activity recognition
- video data