A Dual-Masked Auto-Encoder for Robust Motion Capture with Spatial-Temporal Skeletal Token Completion.
Junkun JiangJie ChenYike GuoPublished in: ACM Multimedia (2022)
Keyphrases
- spatial temporal
- motion capture
- human motion
- human actions
- human body
- spatio temporal
- human motion capture
- human figure
- temporal information
- multi view
- action recognition
- computer graphics
- human computer interaction
- video sequences
- image sequences
- user interface
- spatial and temporal
- high level
- motion estimation
- motion capture data