DEBUG: A Dense Bottom-Up Grounding Approach for Natural Language Video Localization.
Chujie LuLong ChenChilie TanXiaolin LiJun XiaoPublished in: EMNLP/IJCNLP (1) (2019)
Keyphrases
- natural language
- activity detection
- video sequences
- video content
- multimedia
- video data
- dialogue system
- natural language descriptions
- video analysis
- semantic interpretation
- online video
- video streams
- localization algorithm
- knowledge representation
- real time video
- natural language interface
- video images
- semantic representation
- video database
- digital video
- semantic analysis
- real time
- video clips
- video retrieval
- video surveillance
- video frames
- machine learning
- video shots
- natural language generation
- visual saliency
- data driven
- spatio temporal