HSRMamba: Contextual Spatial-Spectral State Space Model for Single Image Hyperspectral Super-Resolution
This work improves hyperspectral imaging for applications like remote sensing by incrementally enhancing a popular model to better handle local and global structural details.
The paper tackles the problem of hyperspectral image super-resolution by addressing Mamba's limitations in capturing spatial-spectral relationships and sensitivity to input order, resulting in a method that outperforms state-of-the-art approaches in quantitative and visual quality.
Mamba has demonstrated exceptional performance in visual tasks due to its powerful global modeling capabilities and linear computational complexity, offering considerable potential in hyperspectral image super-resolution (HSISR). However, in HSISR, Mamba faces challenges as transforming images into 1D sequences neglects the spatial-spectral structural relationships between locally adjacent pixels, and its performance is highly sensitive to input order, which affects the restoration of both spatial and spectral details. In this paper, we propose HSRMamba, a contextual spatial-spectral modeling state space model for HSISR, to address these issues both locally and globally. Specifically, a local spatial-spectral partitioning mechanism is designed to establish patch-wise causal relationships among adjacent pixels in 3D features, mitigating the local forgetting issue. Furthermore, a global spectral reordering strategy based on spectral similarity is employed to enhance the causal representation of similar pixels across both spatial and spectral dimensions. Finally, experimental results demonstrate our HSRMamba outperforms the state-of-the-art methods in quantitative quality and visual results. Code is available at: https://github.com/Tomchenshi/HSRMamba.