Sample Less, Learn More: Efficient Action Recognition via Frame Feature Restoration.
Harry ChengYangyang GuoLiqiang NieZhiyong ChengMohan S. KankanhalliPublished in: CoRR (2023)
Keyphrases
- action recognition
- human actions
- bag of words
- human detection
- computer vision
- activity recognition
- action classification
- motion features
- view invariant
- body parts
- spatial temporal
- image features
- recognizing actions
- max margin
- recognizing human actions
- human pose
- spatio temporal
- efficient learning
- video dataset
- static images
- action detection
- human motion
- feature vectors
- recognition of human actions