Hard No-Box Adversarial Attack on Skeleton-Based Human Action Recognition with Skeleton-Motion-Informed Gradient.
Zhengzhi LuHe WangZiyi ChangGuoan YangHubert P. H. ShumPublished in: CoRR (2023)
Keyphrases
- action recognition
- static images
- human movements
- motion capture data
- motion history images
- recognition of human actions
- recognizing human actions
- human actions
- bag of words
- motion features
- action recognition in videos
- action primitives
- human activities
- activity recognition
- spatial temporal
- action classification
- human detection
- view invariant
- body parts
- computer vision
- mid level
- human motion
- depth sensors
- view invariant action recognition
- image sequences
- video dataset
- human pose
- motion analysis
- camera motion
- human body
- recognizing actions
- bag of features
- motion capture
- motion recognition
- visual words
- motion estimation
- optical flow
- object recognition