Exploring Resolution and Degradation Clues as Self-supervised Signal for Low Quality Object Detection.
Ziteng CuiYingying ZhuLin GuGuo-Jun QiXiaoxiao LiRenrui ZhangZenghui ZhangTatsuya HaradaPublished in: CoRR (2022)
Keyphrases
- low quality
- object detection
- high quality
- signal processing
- high resolution
- face detection
- multi class
- low resolution
- computer vision
- high frequency
- object classification
- poor quality
- pedestrian detection
- object categories
- frequency domain
- non stationary
- super resolution
- coarse to fine
- scene understanding
- object recognition
- scene recognition
- band limited
- original signal