How Incomplete is Contrastive Learning? An Inter-intra Variant Dual Representation Method for Self-supervised Video Recognition.
Lin ZhangQi SheZhengyang ShenChanghu WangPublished in: CoRR (2021)
Keyphrases
- unsupervised learning
- detection method
- learning algorithm
- visual learning
- objective function
- video sequences
- learning process
- prior knowledge
- video data
- neural network
- feature extraction
- active learning
- recognition rate
- supervised learning
- image representation
- reinforcement learning
- feature selection
- segmentation method
- video streams
- learning tasks
- feature representation
- visual recognition
- recognition process
- learning stage