HSRMamba: Contextual Spatial-Spectral State Space Model for Single Image Hyperspectral Super-Resolution

TL;DR

HSRMamba introduces a novel spatial-spectral state space model for hyperspectral image super-resolution, effectively capturing local and global pixel relationships to improve reconstruction quality over existing methods.

Contribution

It proposes a new contextual spatial-spectral modeling approach that addresses limitations of previous models by incorporating patch-wise causal relationships and spectral similarity-based reordering.

Findings

Outperforms state-of-the-art methods in quantitative metrics

Achieves superior visual quality in super-resolved images

Effectively models local and global pixel relationships

Abstract

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.