Cross-Video Contextual Knowledge Exploration and Exploitation for Ambiguity Reduction in Weakly Supervised Temporal Action Localization.
Songchun ZhangChunhui ZhaoPublished in: CoRR (2023)
Keyphrases
- weakly supervised
- contextual knowledge
- object localization
- weakly labeled
- contextual information
- relation extraction
- object class
- topic models
- video sequences
- superpixels
- case based reasoning
- video data
- named entities
- video frames
- object detectors
- semi supervised
- information overload
- information extraction
- domain knowledge
- feature vectors
- feature space
- image segmentation