Decomposed Human Motion Prior for Video Pose Estimation via Adversarial Training.
Wenshuo ChenXiang ZhouZhengdi YuZhaoyu ZhengWeixi GuKai ZhangPublished in: CoRR (2023)
Keyphrases
- pose estimation
- human motion
- human body
- video sequences
- body parts
- human pose
- human actions
- markerless
- motion capture
- motion capture data
- articulated human motion
- visual data
- computer vision
- degrees of freedom
- human movement
- d objects
- human pose estimation
- static images
- human motion capture
- multiple cameras
- video frames
- video data
- particle filtering
- articulated model
- feature points
- upper body
- prior knowledge
- pose parameters
- image sequences
- space time
- articulated objects
- motion tracking
- video surveillance
- motion estimation
- spatio temporal
- training set
- depth images
- human activities
- spatial and temporal
- action recognition