baixuzx7/zerosharpen

Implement of ZeroSharpen

Code for Zero-Sharpen: A universal pansharpening method across satellites for reducing scale-variance gap via zero-shot variation

Configuration requirements before development

1. Python 3.7.0
2. Pytorch 1.12.1

File directory description

ZeroSharpen-main 
├── README.md
├── LICENSE.txt
├── main.py
├── eval.py
├── func.py
├── data
│  ├── ms
│  │    └── 1.tif
│  ├── pan
│  │    └── 1.tif
├── rnet.py
├── tnet.py
└── utils.py

Contributor

Baixuzx7 @ [email protected]

Copyright statement

The project is signed under the MIT license, see the LICENSE.txt

Architecture of Zero-Sharpen

Reference

@article{wang2024zero,
  title={Zero-Sharpen: A universal pansharpening method across satellites for reducing scale-variance gap via zero-shot variation},
  author={Wang, Hebaixu and Zhang, Hao and Tian, Xin and Ma, Jiayi},
  journal={Information Fusion},
  volume={101},
  pages={102003},
  year={2024},
  publisher={Elsevier}
}

zerosharpen's Issues

效果显示问题

您好，非常感谢您提出的先进的全色锐化方法，我有个问题想向您请教，我使用我与您实验设置大小相同的一对图像进行实验，最后得出的实验结果不能在ENVI软件上正常显示，但是我测试了实验结果的部分评价指标是在正常范围内，请问在实验之前除了设置MS与PAN图像的路径之外，还需要调整其他参数吗?感谢您的解答

色域差

作者您好，为何我复现gf2数据集的时候，最后的融合图像色域和gt图相差很大