RCUMP: Residual Completion Unrolling with Mixed Priors for Snapshot Compressive Imaging

Yin-Ping Zhao, Jiancheng Zhang, Yongyong Chen, Zhen Wang, Xuelong Li

This is a repo for our work: "RCUMP: Residual Completion Unrolling with Mixed Priors for Snapshot Compressive Imaging"

News

Our work has been accepted by TIP, codes and results are coming soon (May or June).

Codes and Results

The codes have been released and more details will be updated in a few days.

The simulated and real results of RCUMP are available here.

1. Environment Requirements

Python>=3.6
scipy
numpy

2. Train:

Download the cave dataset of MST series from Baidu diskcode:fo0q or here, put the dataset into the corresponding folder "RCUMP/CAVE_1024_28/" as the following form:

|--CAVE_1024_28
    |--scene1.mat
    |--scene2.mat
    ：
    |--scene205.mat
    |--train_list.txt

Then run the following command

cd RCUMP
python Train.py

3. Test:

Download the test dataset from here, put the dataset into the corresponding folder "RCUMP/Test_data/" as the following form:

|--Test_data
    |--scene01.mat
    |--scene02.mat
    ：
    |--scene10.mat
    |--test_list.txt

Then run the following command

cd RCUMP
python Test.py

Finally, run 'cal_psnr_ssim.m' in Matlab to get the performance metrics.

Citation

If this repo helps you, please consider citing our work:

@ARTICLE{RCUMP,
  author={Zhao, Yin-Ping and Zhang, Jiancheng and Chen, Yongyong and Wang, Zhen and Li, Xuelong},
  journal={IEEE Transactions on Image Processing}, 
  title={RCUMP: Residual Completion Unrolling With Mixed Priors for Snapshot Compressive Imaging}, 
  year={2024},
  volume={33},
  number={},
  pages={2347-2360},
  keywords={Imaging;Image coding;Iterative methods;Optimization;Image reconstruction;Hyperspectral imaging;Artificial neural networks;Snapshot compressive imaging;hyperspectral image;deep unrolling-based methods},
  doi={10.1109/TIP.2024.3374093}}