PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action Recognition.
Neel TrivediRavi Kiran SarvadevabhatlaPublished in: CoRR (2022)
Keyphrases
- action recognition
- human pose
- recognizing actions
- human actions
- bag of words
- activity recognition
- spatial temporal
- body parts
- pose estimation
- human detection
- computer vision
- action classification
- view invariant action recognition
- multi modal
- recognition of human actions
- partial occlusion
- action detection
- recognizing human actions
- d objects
- viewpoint
- machine learning