Xiangyu Rui¹, Xiangyong Cao¹, Qi Xie¹, Zongsheng Yue¹, Qian Zhao¹, Deyu Meng^1,2

¹Xi’an Jiaotong University; ²Pazhou Lab, Guangzhou

Main paper

Supplement material

1. python >= 3.8

2. pytorch = 1.9 (lower version may also be applicable.)

1. Please download CAVE DATAset from https://www.cs.columbia.edu/CAVE/databases/multispectral/ for training. The image size is of 512*512*31.

2. Crop training dataset. I randomly select 20 images for training and crop about 10000 patches of size 96*96*20. The corresponding MATLAB codes are in "data/gene_patches.m". (You could also freely choose your favourite way to crop patches.)

3. Save the training dataset in path "dataroot"(your own data path).

Please refer to "data/gene_test_data.m" file for creating your own test data using MATLAB. The noise generation methods in "data/utils" file are in consistent with those in "lib.py".

Plean refer to "train_hwnet.py" and "test.py" for training and testing HWLRMF. More test codes for NAILRMA, NGmeet, LLRT and their weighted versions will be uploaded soon.

For pytorch>=1.9, torch.linalg.svd could also be directly used. However, sometimes the grads could be numerically unstable.

If you are interested in our work, please cite

@InProceedings{Rui_2021_CVPR, 
    author    = {Rui, Xiangyu and Cao, Xiangyong and Xie, Qi and Yue, Zongsheng and Zhao, Qian and Meng, Deyu},
    title     = {Learning an Explicit Weighting Scheme for Adapting Complex HSI Noise},    
    booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},    
    month     = {June},    
    year      = {2021},    
    pages     = {6739-6748}    
}

If you have any questions, please contract me by [email protected] or [email protected].