ConTra: (Con)text (Tra)nsformer for Cross-Modal Video Retrieval.
Adriano FragomeniMichael WrayDima DamenPublished in: CoRR (2022)
Keyphrases
- video retrieval
- cross modal
- video search
- concept based video retrieval
- multi modal
- content based retrieval
- multimedia retrieval
- visual content
- semantic gap
- visual data
- video data
- retrieval systems
- text retrieval
- visual similarity
- information retrieval
- multimedia documents
- video content
- key frames
- image retrieval
- multimedia databases
- text mining
- semantic content
- object recognition
- keywords
- semantic information
- image data
- metadata