A part of this dataset was used in the RealSR challenge in NTIRE 2019 (in conjunction with CVPR 2019).

Detail for training & testing: Trained on the RGB domain and tested on Y channel (images from Version 3).

Visualization (zooming factor: 4)
More results can be downloaded here.

1. Models for PSNR/SSIM
2. Models for Visualization

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.

1. Download the new layers in folder 'Layer'
2. Modify the caffe.proto (Path: caffe/src/caffe/proto/)
3. Compile Caffe and Matcaffe (installation)

-- Training --

4. Generate the training data
5. run *solver.prototxt to train the network

-- Testing --

6. run Test.m

1. Put your own image pairs in the folder and modify the path
2. run Demo.m in folder 'Alignment'
3. 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: [email protected]).