Triadic temporal-semantic alignment for weakly-supervised video moment retrieval.
Jin LiuJialong XieFengyu ZhouShengfeng HePublished in: Pattern Recognit. (2024)
Keyphrases
- weakly supervised
- semantic attributes
- weakly labeled
- topic models
- superpixels
- relation extraction
- video data
- information retrieval
- object class
- object detectors
- semi supervised
- video sequences
- video frames
- image retrieval
- image database
- semantic features
- semantic information
- domain specific
- low level features
- object detection
- high level
- named entities
- graph cuts
- relevance feedback