VELOCITI: Can Video-Language Models Bind Semantic Concepts through Time?
Darshana SaravananDarshan Singh SVarun GuptaZeeshan KhanVineet GandhiMakarand TapaswiPublished in: CoRR (2024)
Keyphrases
- language model
- semantic concepts
- semantic concept detection
- multi modal
- video analysis
- visual features
- semantic similarity
- probabilistic model
- low level
- n gram
- low level features
- multimedia content
- multimedia data
- visual content
- information retrieval
- video shots
- query expansion
- key frames
- retrieval model
- image annotation
- image collections
- test collection
- semantic information
- video data
- generative model
- video retrieval
- higher level
- co occurrence
- domain knowledge
- video sequences
- video content
- visual information
- image classification
- video database
- multimedia