Are Binary Annotations Sufficient? Video Moment Retrieval via Hierarchical Uncertainty-based Active Learning.
Wei JiRenjie LiangZhedong ZhengWenqiao ZhangShengyu ZhangJuncheng LiMengze LiTat-Seng ChuaPublished in: CVPR (2023)
Keyphrases
- active learning
- content based indexing
- relevance feedback
- video data
- multimedia information
- content based video
- video indexing
- video content
- semantic video
- video sequences
- image retrieval
- multimedia
- audio visual content
- key frames
- selective sampling
- cut detection
- test collection
- video frames
- video streams
- semantic concept detection
- information retrieval
- video search
- video database
- natural language descriptions
- web image retrieval
- multimedia search
- video material
- textual descriptions
- video clips
- multimedia documents
- multimedia databases
- image database
- semantic annotation
- labeled data
- visual concepts
- metadata
- video retrieval
- query expansion
- news video
- video dataset
- document retrieval
- random sampling
- learning algorithm
- keywords
- semi supervised
- multimedia data
- image annotation
- real time
- user generated
- semantic labels
- text queries
- visual similarity
- information retrieval systems
- automatic annotation
- training set
- content based retrieval
- machine learning