Source-Free Video Domain Adaptation by Learning Temporal Consistency for Action Recognition.
Yuecong XuJianfei YangHaozhi CaoKeyu WuMin WuZhenghua ChenPublished in: ECCV (34) (2022)
Keyphrases
- action recognition
- domain adaptation
- temporal consistency
- human actions
- learning algorithm
- multiple sources
- computer vision
- action classification
- target domain
- learning tasks
- recognizing human actions
- cross domain
- video frames
- labeled data
- active learning
- three dimensional
- test data
- text categorization
- human activities
- temporal information
- key frames
- semi supervised
- image features
- high resolution
- video sequences
- keywords
- decision trees