Rethinking matching-based few-shot action recognition.
Juliette BertrandYannis KalantidisGiorgos ToliasPublished in: CoRR (2023)
Keyphrases
- action recognition
- human actions
- bag of words
- human detection
- activity recognition
- body parts
- computer vision
- depth sensors
- recognizing human actions
- action classification
- recognition of human actions
- image matching
- spatial temporal
- mid level
- action primitives
- feature points
- video sequences
- motion history images
- action recognition in videos
- independent subspace analysis
- view invariant action recognition
- video dataset
- static images
- bag of features
- video data
- low level
- depth cameras
- human activities
- visual features
- spatio temporal
- object recognition
- high level