CLIP-guided Prototype Modulating for Few-shot Action Recognition.
Xiang WangShiwei ZhangJun CenChangxin GaoYingya ZhangDeli ZhaoNong SangPublished in: Int. J. Comput. Vis. (2024)
Keyphrases
- action recognition
- bag of words
- human actions
- computer vision
- activity recognition
- human detection
- action classification
- key frames
- recognizing actions
- body parts
- bag of features
- video data
- visual features
- static images
- video sequences
- video clips
- recognition of human actions
- recognizing human actions
- spatial temporal
- view invariant
- independent subspace analysis
- mid level
- video shots
- video content
- video dataset
- spatio temporal
- depth sensors
- action primitives
- detection algorithm