Disentangling and Unifying Graph Convolutions for Skeleton-Based Action Recognition.
Ziyu LiuHongwen ZhangZhenghao ChenZhiyong WangWanli OuyangPublished in: CoRR (2020)
Keyphrases
- action recognition
- bag of words
- human actions
- activity recognition
- human detection
- action classification
- spatial temporal
- body parts
- recognition of human actions
- recognizing actions
- independent subspace analysis
- static images
- computer vision
- human activities
- max margin
- depth sensors
- view invariant
- mid level
- bag of features
- graph structure
- image classification
- space time interest points
- view invariant action recognition
- human pose
- motion history images
- human activity recognition
- video dataset
- pose estimation
- multi class
- probabilistic model