UniFormerV2: Unlocking the Potential of Image ViTs for Video Understanding.
Kunchang LiYali WangYinan HeYizhuo LiYi WangLimin WangYu QiaoPublished in: ICCV (2023)
Keyphrases
- image data
- single image
- image classification
- image features
- image representation
- input image
- static images
- image retrieval
- image collections
- test images
- image content
- high resolution
- multiscale
- low level
- image analysis
- video data
- image regions
- multimedia
- region of interest
- image pixels
- object motion
- video images
- image frames
- visual cues
- video files
- template matching
- segmentation method
- edge detection
- image segmentation
- video content
- pixel values
- key frames
- feature points
- video sequences
- pre trained
- images and video sequences