Multi-level Dispersion Residual Network for Efficient Image Super-Resolution [paper]
Yanyu Mao1, Nihao Zhang1, Qian Wang2, Bendu Bai, Wanying Bai, Haonan Fang, Peng Liu, Mingyue Li, Shengbo Yan
pip install -r requirements.txt -i http://pypi.douban.com/simple/ --trusted-host pypi.douban.com
The trainset uses the DIV2K (800) + LSDIR(the first 10k). In order to effectively improve the training speed, images are cropped to 480 * 480 images by running script extract_subimages.py, and the dataloader will further randomly crop the images to the GT_size required for training. GT_size defaults to 128/192/256 (×2/×3/×4).
python extract_subimages.py
The input and output paths of cropped pictures can be modify in this script. Default location: ./datasets/DL2K.
### Train ###
### MDRN ###
python train.py -opt ./options/train/MDRN/train_mdrn_x2.yml --auto_resume # ×2
python train.py -opt ./options/train/MDRN/train_mdrn_x3.yml --auto_resume # ×3
python train.py -opt ./options/train/MDRN/train_mdrn_x4.yml --auto_resume # ×4
For more training commands, please check the docs in BasicSR
### Test ###
### MDRN ###
python basicsr/test.py -opt ./options/test/MDRN/test_mdrn_x2.yml # ×2
python basicsr/test.py -opt ./options/test/MDRN/test_mdrn_x3.yml # ×3
python basicsr/test.py -opt ./options/test/MDRN/test_mdrn_x4.yml # ×4
The inference results on benchmark datasets are available at Google Drive or Baidu Netdisk (access code: TeLX).
If you have any questions, please feel free to contact us, zwyczhang@stu.xupt.edu.cn and bolttt@stu.xupt.edu.cn.
@InProceedings{Mao_2023_CVPR,
author = {Mao, Yanyu and Zhang, Nihao and Wang, Qian and Bai, Bendu and Bai, Wanying and Fang, Haonan and Liu, Peng and Li, Mingyue and Yan, Shengbo},
title = {Multi-Level Dispersion Residual Network for Efficient Image Super-Resolution},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
month = {June},
year = {2023},
pages = {1660-1669}
}