Baisonm-Li / HSR-KAN

The code of "HSR-KAN: Hyperspectral Image Super-Resolution based on Kolmogorov-Arnold Networks"
12 stars 1 forks source link

HSR-KAN

The code of "HSR-KAN: Efficient Hyperspectral Image Super-Resolution via Kolmogorov-Arnold Networks" [arxiv]

Hyperspectral images (HSIs) have great potential in various visual tasks due to their rich spectral information. However, obtaining high-resolution hyperspectral images remains challenging due to limitations of physical imaging. Inspired by Kolmogorov-Arnold Networks(KANs), we propose efficient HSI super-resolution (HSI-SR) model to fuse a low-resolution HSI (LR-HSI) and a high-resolution multispectral image (HR-MSI), yielding a high-resolution HSI (HR-HSI). To achieve the effective integration of spatial information from HR-MSI, we design a fusion module based on KANs, called KAN-Fusion. Further inspired by the channel attention mechanism, we design a spectral channel attention module called KAN Channel Attention Block (KAN-CAB) for post-fusion feature extraction. As a channel attention module integrated with KANs, KAN-CAB not only enhances the fine-grained adjustment ability of deep networks, enabling networks to accurately simulate details of spectral sequences and spatial textures, but also effectively avoid Curse of Dimensionality (COD). Extensive experiments show that, compared to current state-of-the-art (SOTA) HSI-SR methods, proposed HSR-KAN achieves the best performance in terms of both qualitative and quantitative assessments.

Requirements

Datasets

Usage

python train.py --dataset CAVE --batch_size 32 --epochs 1000

We publish pre-trained network weights in the models folder, which you can load and use directly!

Experiments Results

x2,x4,x8 SR on CAVE,Chikusei,Havard datasets

alt text

Computational Efficiency

alt text

Ablation experiments

alt text

Acknowledgement

Our code references efficient-kan and pykan. Thanks for their greak work!