Domain Adaptation in Multi-View Embedding for Cross-Modal Video Retrieval.
Jonathan MunroMichael WrayDiane LarlusGabriela CsurkaDima DamenPublished in: CoRR (2021)
Keyphrases
- video retrieval
- multi view
- co training
- semi supervised
- single view
- labeled data
- visual content
- multiple views
- semantic gap
- video data
- unlabeled data
- d objects
- three dimensional
- retrieval systems
- vector space
- key frames
- multi modal
- video content
- semi supervised learning
- training data
- similarity search
- image classification
- transfer learning
- text mining
- visual features
- computer vision
- active learning
- pairwise
- visual data