TimeBalance: Temporally-Invariant and Temporally-Distinctive Video Representations for Semi-Supervised Action Recognition.
Ishan Rajendrakumar DaveMamshad Nayeem RizveChen ChenMubarak ShahPublished in: CVPR (2023)
Keyphrases
- action recognition
- human actions
- action classification
- spatial temporal
- mid level
- semi supervised
- video dataset
- spatio temporal
- temporal information
- action detection
- recognizing human actions
- recognition of human actions
- motion features
- bag of words
- static images
- human activities
- computer vision
- space time interest points
- human detection
- video sequences
- activity recognition
- body parts
- recognizing actions
- video analysis
- depth sensors
- space time
- video frames
- multi view
- action primitives
- key frames
- video streams
- view invariant
- video surveillance
- human pose
- motion history images
- video clips
- semi supervised learning
- bag of features
- labeled data
- action recognition in videos
- video shots