In Defense of Image Pre-Training for Spatiotemporal Recognition.
Xianhang LiHuiyu WangChen WeiJieru MeiAlan L. YuilleYuyin ZhouCihang XiePublished in: CoRR (2022)
Keyphrases
- image data
- single image
- input image
- multiscale
- image content
- image matching
- object recognition
- image representation
- image features
- image classification
- image segmentation
- segmentation method
- static images
- high resolution
- image analysis
- spatial information
- image regions
- keypoints
- region of interest
- image collections
- segmentation algorithm
- improve recognition accuracy
- intrusion detection
- feature points
- handwritten words
- test images
- object models
- image pixels
- handwriting recognition
- recognition scheme
- action recognition
- recognition algorithm
- hough transform
- binary images
- super resolution
- object detection
- video sequences
- training data
- similarity measure