SVT: Supertoken Video Transformer for Efficient Video Understanding.
Chenbin PanRui HouHanchao YuQifan WangSenem VelipasalarMadian KhabsaPublished in: CoRR (2023)
Keyphrases
- video sequences
- video data
- multimedia
- video content
- real time
- video streams
- space time
- video analysis
- digital video
- video frames
- spatial and temporal
- video clips
- event recognition
- artificial intelligence
- video database
- surveillance videos
- online video
- visual data
- video retrieval
- computationally efficient
- low level
- expert systems
- computer vision
- neural network