MultiVENT: Multilingual Videos of Events with Aligned Natural Text.
Kate SandersDavid EtterReno KrizBenjamin Van DurmePublished in: CoRR (2023)
Keyphrases
- event recognition
- video event
- text generation
- human activities
- event detection
- multi lingual
- video clips
- information retrieval
- video search
- video sequences
- temporal relationships
- natural language descriptions
- video dataset
- language independent
- news video
- spatio temporal patterns
- video data
- text mining
- keywords
- video segments
- text retrieval
- video event detection
- unusual events
- video frames
- video collections
- event models
- video database
- tv news
- natural language generation
- surveillance videos
- video surveillance
- temporal information
- temporal events
- multilingual documents
- text documents
- action recognition
- caption text
- complex events
- temporal structure
- textual descriptions
- temporal relations
- news stories
- cross language information retrieval
- video analysis
- machine translation
- digital libraries