SVFormer: Semi-supervised Video Transformer for Action Recognition.
Zhen XingQi DaiHan HuJingjing ChenZuxuan WuYu-Gang JiangPublished in: CoRR (2022)
Keyphrases
- action recognition
- human actions
- semi supervised
- action classification
- spatial temporal
- action detection
- video dataset
- motion features
- static images
- recognizing human actions
- bag of words
- recognition of human actions
- computer vision
- activity recognition
- human detection
- space time interest points
- human activities
- view invariant
- video sequences
- mid level
- motion history images
- multimedia
- bag of features
- body parts
- video streams
- video data
- recognizing actions
- pairwise
- active learning
- human pose
- motion capture data
- video analysis
- depth sensors
- object detection
- video shots
- atomic actions