Multiscale RGB-Guided Fusion for Hyperspectral Image Super-Resolution

TL;DR

This paper introduces CGNet, a new method that improves the spatial resolution of hyperspectral images using high-resolution RGB images while maintaining spectral accuracy.

Contribution

The novel contribution is a dual-encoder network with a multi-scale fusion decoder and a hybrid loss function for hyperspectral image super-resolution.

Findings

CGNet outperforms existing methods on the ARAD1K and StereoMSI benchmarks in terms of PSNR and SSIM.

The hybrid loss combining L1 and SAM is shown to be the most effective formulation through ablation studies.

CGNet achieves lower spectral angle mapper (SAM) and ΔE00 values, indicating better spectral fidelity and color accuracy.

Abstract

Hyperspectral imaging (HSI) enables fine spectral analysis but is often limited by low spatial resolution due to sensor constraints. To address this, we propose CGNet, a color-guided hyperspectral super-resolution network that leverages complementary information from low-resolution hyperspectral inputs and high-resolution RGB images. CGNet adopts a dual-encoder design: the RGB encoder extracts hierarchical spatial features, while the HSI encoder progressively upsamples spectral features. A multi-scale fusion decoder then combines both modalities in a coarse-to-fine manner to reconstruct the high-resolution HSI. Training is driven by a hybrid loss that balances L1 and Spectral Angle Mapper (SAM), which ablation studies confirm as the most effective formulation. Experiments on two benchmarks, ARAD1K and StereoMSI, at ×4 and ×6 upscaling factors demonstrate that CGNet consistently outperforms state-of-the-art baselines. CGNet achieves higher PSNR and SSIM, lower SAM, and reduced ΔE00, confirming its ability to recover sharp spatial structures while preserving spectral fidelity.

Figures