Home

Awesome

QRNN3D

The implementation of TNNLS 2020 paper "3D Quasi-Recurrent Neural Network for Hyperspectral Image Denoising"

🌟 See also the follow up works of QRNN3D:

📉 Performance: QRNN3D < DPHSIR < MAN < HSDT

Highlights

<img src="imgs/runtime_gauss.png" height="140px"/> <img src="imgs/runtime_complex.png" height="140px"/>

<img src="imgs/PaviaU.gif" height="140px"/> <img src="imgs/Indian_pines.gif" height="140px"/> <img src="imgs/Urban.gif" height="140px"/>

Prerequisites

Quick Start

1. Preparing your training/testing datasets

Download ICVL hyperspectral image database from here (we only need .mat version)

Training dataset

Note cafe (via conda install) and lmdb are required to execute the following instructions.

Testing dataset

Note matlab is required to execute the following instructions.

2. Testing with pretrained models

You can also use hsi_eval.py to evaluate quantitative HSI denoising performance.

3. Training from scratch

Citation

If you find this work useful for your research, please cite:

@article{wei2020QRNN3D,
  title={3-D Quasi-Recurrent Neural Network for Hyperspectral Image Denoising},
  author={Wei, Kaixuan and Fu, Ying and Huang, Hua},
  journal={IEEE Transactions on Neural Networks and Learning Systems},
  year={2020},
  publisher={IEEE}
}

and follow up works

@article{lai2022dphsir,
    title = {Deep plug-and-play prior for hyperspectral image restoration},
    journal = {Neurocomputing},
    volume = {481},
    pages = {281-293},
    year = {2022},
    issn = {0925-2312},
    doi = {https://doi.org/10.1016/j.neucom.2022.01.057},
    author = {Zeqiang Lai and Kaixuan Wei and Ying Fu},
}

@inproceedings{lai2023hsdt,
  author = {Lai, Zeqiang and Chenggang, Yan and Fu, Ying},
  title = {Hybrid Spectral Denoising Transformer with Guided Attention},
  booktitle={Proceedings of the IEEE International Conference on Computer Vision},
  year = {2023},
}

@article{lai2023mixed,
  title={Mixed Attention Network for Hyperspectral Image Denoising},
  author={Lai, Zeqiang and Fu, Ying},
  journal={arXiv preprint arXiv:2301.11525},
  year={2023}
}

Contact

Please contact me if there is any question (kxwei@princeton.edu kaixuan_wei@bit.edu.cn)