GF6SRD: a real-world cross-sensor remote sensing super-resolution dataset

We used cross-sensor satellite imagery to construct the real-world cross-sensor remote sensing dataset, GF6SRD. This dataset is proposed in the paper "DESAT: A Distance Enhanced Strip Attention Transformer for Remote Sensing Super-Resolution"

The code of this paper is also available at https://github.com/afomyj/DESAT

If you find our work helpful, please cite:

Mao Y, He G, Wang G, Yin R, Peng Y, Guan B. DESAT: A Distance-Enhanced Strip Attention Transformer for Remote Sensing Image Super-Resolution. Remote Sensing. 2024; 16(22):4251. https://doi.org/10.3390/rs16224251

The exposed portion of this resource is part of GF6SRD, containg 2000 LR-HR image pairs for training and 500 pairs for testing. The full GF6SRD dataset can be obtained under appropriate request to the coresponding author, He, Guojin. Email: [e-mail]