HyRSM++: Hybrid Relation Guided Temporal Set Matching for Few-shot Action Recognition.
Xiang WangShiwei ZhangZhiwu QingZhengrong ZuoChangxin GaoRong JinNong SangPublished in: CoRR (2023)
Keyphrases
- action recognition
- spatial temporal
- human actions
- human detection
- computer vision
- bag of words
- video sequences
- recognizing human actions
- action classification
- static images
- body parts
- activity recognition
- temporal information
- spatial and temporal
- video dataset
- space time
- action detection
- low level
- view invariant action recognition