Raw images can be downloaded HERE
A part of this dataset was used in the RealSR challenge in NTIRE 2019 (in conjunction with CVPR 2019).
Version 1: 234 scenes (204 scenes for training & 30 scenes for testing), as reported in the original paper (HR has the same resolution as LR).
Version 2: 559 scenes (459 scenes for training & 100 scenes for testing), the extended version (HR has the same resolution as LR).
Methods | PSNR | 2 | 3 | 4 | SSIM | 2 | 3 | 4 |
---|---|---|---|---|---|---|---|---|
KPN(K=5) | 33.41 | 30.47 | 28.80 | 0.913 | 0.860 | 0.826 | ||
KPN(K=7) | 33.42 | 30.49 | 28.84 | 0.913 | 0.861 | 0.826 | ||
KPN(K=13) | 33.44 | 30.52 | 28.92 | 0.913 | 0.863 | 0.829 | ||
KPN(K=19) | 33.45 | 30.57 | 28.99 | 0.914 | 0.864 | 0.832 | ||
LP-KPN(K=5) | 33.49 | 30.60 | 29.05 | 0.917 | 0.865 | 0.834 |
Version 3: 559 scenes (459 scenes for training & 100 scenes for testing), the extended version (HR and LR have different resolution).
Detail for training & testing: Trained on the RGB domain and tested on Y channel (images from Version 3).
Methods | PSNR | 2 | 3 | 4 | SSIM | 2 | 3 | 4 |
---|---|---|---|---|---|---|---|---|
Bicubic | - | 28.6284 | 27.2378 | - | 0.8088 | 0.7643 | ||
Baseline (Our) | - | 30.6003 | 28.6508 | - | 0.8630 | 0.8206 |
Visualization (zooming factor: 4)
More results can be downloaded here.
The above provided models (for quantitative metrics and visual quality) are both trained with the loss ratio in 1:1:1.
We select different models at different epochs for different purposes.
- Download the new layers in folder 'Layer'
- Modify the caffe.proto (Path: caffe/src/caffe/proto/)
- Compile Caffe and Matcaffe (installation)
-- Training --
- Generate the training data
- run *solver.prototxt to train the network
-- Testing --
- run Test.m
- Put your own image pairs in the folder and modify the path
- run Demo.m in folder 'Alignment'
- Central region crop
(1) coarse align the image pairs;
(2) central crop the image pairs;
(3) finer align the cropped image pairs;
(4) discard those misaligned image pairs.
If you find this work useful for your research, please cite:
@inproceedings{cai2019toward,
title={Toward real-world single image super-resolution: A new benchmark and a new model},
author={Cai, Jianrui and Zeng, Hui and Yong, Hongwei and Cao, Zisheng and Zhang, Lei},
booktitle={Proceedings of the IEEE International Conference on Computer Vision},
year={2019}
}
@inproceedings{cai2019ntire,
title={Ntire 2019 challenge on real image super-resolution: Methods and results},
author={Cai, Jianrui and Gu, Shuhang and Timofte, Radu and Zhang, Lei},
booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops},
year={2019}
}
Please contact me if there is any question (Jianrui CAI: csjcai@comp.polyu.edu.hk).