Home

Awesome

DDS2M: Self-Supervised Denoising Diffusion Spatio-Spectral Model for Hyperspectral Image Restoration

This repository contains the code for the paper

[ICCV 2023] DDS2M: Self-Supervised Denoising Diffusion Spatio-Spectral Model for Hyperspectral Image Restoration
Yuchun Miao, Lefei Zhang, Liangpei Zhang, Dacheng Tao

<div align="center"> <img src="figures/motivation.png" width="500px" /> </div>

Installation

Clone this repository:

git clone git@github.com:miaoyuchun/DDS2M.git

The project was developed using Python 3.7.10, and torch 1.12.1. You can build the environment via pip as follow:

pip3 install -r requirements.txt

Running Experiments

We provide code to reproduce the main results on HSI completion, HSI denoising, and HSI super-resolution as follows:

python main_completion.py
python main_denoising.py
python main_sisr.py

Citation and Acknowledgement

If you find our work useful in your research, please cite:

@article{miao2023dds2m,
  title={DDS2M: Self-Supervised Denoising Diffusion Spatio-Spectral Model for Hyperspectral Image Restoration},
  author={Miao, Yuchun and Zhang, Lefei and Zhang, Liangpei and Tao, Dacheng},
  journal={arXiv preprint arXiv:2303.06682},
  year={2023}
}

The code is highly based on the repository of DS2DP, DDRM, and DDPM.