[IEEE TGRS 2024] EDiffSR: 一个高效的扩散概率模型,用于遥感图像超分辨率
PyTorch codes for "EDiffSR: An Efficient Diffusion Probabilistic Model for Remote Sensing Image Super-Resolution", IEEE Transactions on Geoscience and Remote Sensing, 2024.
Authors: Yi Xiao, Qiangqiang Yuan*, Kui Jiang, Jiang He, Xianyu Jin, and Liangpei Zhang
Wuhan University and Harbin Institute of Technology
Step I. Please download the following remote sensing benchmarks:
| Data Type | AID | DOTA-v1.0 | DIOR | NWPU-RESISC45 |
|---|---|---|---|---|
| Training | Download | None | None | None |
| Testing | Download | Download | Download | Download |
Please refer to Dataset Processing to build the LR-HR training pairs.
Step II. Modify the path in options/train/setting.yml and options/test/aid.yml
python train.py -opt=options/train/setting.ymlpython test.py -opt=options/test/nwpu.ymlOur EDiffSR mainly borrows from the SDE process of IRSDE and NAFNet. Thanks for these excellent open-source works!
If you have any questions or suggestions, feel free to contact me.
Email: [email protected]; [email protected]
If our work contributes to your research, we would greatly appreciate your citation. Thank you for your support!
@ARTICLE{xiao2024ediffsr,
author={Xiao, Yi and Yuan, Qiangqiang and Jiang, Kui and He, Jiang and Jin, Xianyu and Zhang, Liangpei},
journal={IEEE Transactions on Geoscience and Remote Sensing},
title={EDiffSR: An Efficient Diffusion Probabilistic Model for Remote Sensing Image Super-Resolution},
year={2024},
volume={62},
number={},
pages={1-14},
doi={10.1109/TGRS.2023.3341437}
}暂无开放 Issues,或尚未同步最近议题。