CLIP-Lite: Information Efficient Visual Representation Learning from Textual Annotations.
Aman ShrivastavaRamprasaath R. SelvarajuNikhil NaikVicente OrdonezPublished in: CoRR (2021)
Keyphrases
- visual representation
- keywords
- prior knowledge
- learning process
- learning systems
- information sources
- learning algorithm
- background knowledge
- metadata
- visual representations
- learning tasks
- online learning
- domain knowledge
- knowledge acquisition
- supervised learning
- learning design
- video sequences
- reinforcement learning
- scientific visualization