A Closer Look at Debiased Temporal Sentence Grounding in Videos: Dataset, Metric, and Approach.
Xiaohan LanYitian YuanXin WangLong ChenZhi WangLin MaWenwu ZhuPublished in: ACM Trans. Multim. Comput. Commun. Appl. (2023)
Keyphrases
- human actions
- video sequences
- temporal coherence
- temporal domain
- temporal relationships
- action recognition
- temporal information
- temporal reasoning
- multimedia event detection
- video database
- temporal data
- natural language
- event recognition
- temporal constraints
- spatial and temporal
- distance measure
- dynamic textures
- video dataset
- temporal segmentation
- space time
- trecvid multimedia event detection
- metric space
- web videos
- weakly labeled
- temporal structure
- spatio temporal
- human activities
- feature selection
- motion trajectories
- benchmark datasets
- temporal patterns
- video frames
- event detection
- semantic role labeling
- part of speech
- temporal consistency
- video data
- million images
- personal photos
- video analysis
- evaluation metrics