RaP: Redundancy-aware Video-language Pre-training for Text-Video Retrieval.
Xing WuChaochen GaoZijia LinZhongyuan WangJizhong HanSonglin HuPublished in: EMNLP (Findings) (2022)
Keyphrases
- video retrieval
- video search
- video segments
- video collections
- video database
- concept based video retrieval
- video data
- video content
- video indexing
- content based video retrieval
- content based retrieval
- key frames
- video clips
- video shots
- visual content
- semantic gap
- concept detection
- information retrieval
- news video
- retrieval systems
- semantic concept detection
- video summarization
- video sequences
- video streams
- digital video
- semantic content
- keywords
- broadcast news
- semantic video
- face recognition
- video analysis
- natural language
- high level
- multi modal