Stitching Segments and Sentences towards Generalization in Video-Text Pre-training.
Fan MaXiaojie JinHeng WangJingjia HuangLinchao ZhuYi YangPublished in: AAAI (2024)
Keyphrases
- video segments
- sentence level
- natural language descriptions
- training corpus
- video data
- video sequences
- multimedia
- natural language
- text segments
- text generation
- text detection
- video content
- lexical features
- video retrieval
- video clips
- text corpus
- long video
- information retrieval
- video streams
- text summarization
- video search
- text mining
- extractive summarization
- human generated
- syntactic analysis
- learning machines
- multimedia documents
- training set
- text documents
- video analysis
- syntactic structures
- linguistic analysis
- discourse structure
- semantic representations
- automatic summarization
- syntactic information
- video database
- news video
- event detection
- temporal segmentation
- visual information
- semantic information
- dependency tree
- sentence similarity
- language model