Awesome
CloserLookBlindSR
Paper accepted by CVPR Workshop, NTIRE 2022
A Closer Look at Blind Super-Resolution: Degradation Models, Baselines, and Performance Upper Bounds
By Wenlong Zhang, Guangyuan Shi, Yihao Liu, Chao Dong, Xiao-Ming Wu
<p align="center"> <img height="270" src="./figure/method.png"> </p>An example of the gate operation on adding blur
############## add blur ###################
# self.opt['gate_blur_prob'] = 0.5
if np.random.uniform() < self.opt['gate_blur_prob']:
out = filter2D(self.gt, self.kernel1)
else:
out = self.gt
Citation
Please cite us if our work is useful for your research.
@InProceedings{Zhang_2022_CVPR,
author = {Zhang, Wenlong and Shi, Guangyuan and Liu, Yihao and Dong, Chao and Wu, Xiao-Ming},
title = {A Closer Look at Blind Super-Resolution: Degradation Models, Baselines, and Performance Upper Bounds},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
month = {June},
year = {2022},
pages = {527-536}
}
License
This project is released under the Apache 2.0 license.
Acknowledgement
The code is based on BasicSR and Real-ESRGAN. Thanks to Xintao Wang.
Contact
If you have any questions, please feel free to contact with me at zhangwenlong0517@gmail.com