Self-supervised pretext task collaborative multi-view contrastive learning for video action recognition.
Shuai BiZhengping HuMengyao ZhaoHehao ZhangJirui DiZhe SunPublished in: Signal Image Video Process. (2023)
Keyphrases
- multi view
- action recognition
- view invariant
- multi view learning
- action classification
- human actions
- multiple views
- single view
- d objects
- dynamic scenes
- free viewpoint
- learning process
- depth map
- computer vision
- surface reconstruction
- multiple cameras
- human activities
- activity recognition
- learning algorithm
- action detection
- learning tasks
- statistical learning
- motion capture
- visual hull
- active learning
- object recognition
- recognizing human actions