Expanding Language-Image Pretrained Models for General Video Recognition.
Bolin NiHouwen PengMinghao ChenSongyang ZhangGaofeng MengJianlong FuShiming XiangHaibin LingPublished in: ECCV (4) (2022)
Keyphrases
- object models
- static images
- image data
- image matching
- image content
- image analysis
- input image
- random fields
- image representation
- bayesian framework
- image frames
- object recognition
- image retrieval
- image collections
- single image
- image features
- template matching
- multiscale
- three dimensional objects
- recognition rate
- video images
- preprocessing stage
- edge detection
- probabilistic model
- image segmentation
- multimedia
- image classification
- test images
- low level
- partial occlusion
- temporal continuity
- road signs
- computer vision
- key frames
- human activities
- activity recognition
- spatial information
- keypoints
- action recognition
- high resolution
- moving objects