Towards Understanding the Adversarial Vulnerability of Skeleton-based Action Recognition.
Tianhang ZhengSheng LiuChangyou ChenJunsong YuanBaochun LiKui RenPublished in: CoRR (2020)
Keyphrases
- action recognition
- bag of words
- human detection
- action classification
- human actions
- activity recognition
- computer vision
- body parts
- spatial temporal
- static images
- mid level
- recognizing human actions
- bag of features
- human activities
- recognition of human actions
- recognizing actions
- video dataset
- view invariant
- action detection
- independent subspace analysis
- action recognition in videos
- space time interest points
- action primitives
- motion features
- max margin
- human pose
- multiscale