GEWDiff: Geometric Enhanced Wavelet-based Diffusion Model for Hyperspectral Image Super-resolution

TL;DR

GEWDiff is a novel hyperspectral image super-resolution framework that combines wavelet-based compression, geometry-aware diffusion, and multi-level loss to achieve state-of-the-art reconstruction quality.

Contribution

It introduces a wavelet encoder-decoder and a geometry-enhanced diffusion process tailored for hyperspectral images, addressing high dimensionality and geometric preservation challenges.

Findings

Achieved 4x super-resolution of hyperspectral images.

Demonstrated superior spectral and visual fidelity over existing methods.

Ensured stable convergence with a multi-level loss function.

Abstract

Improving the quality of hyperspectral images (HSIs), such as through super-resolution, is a crucial research area. However, generative modeling for HSIs presents several challenges. Due to their high spectral dimensionality, HSIs are too memory-intensive for direct input into conventional diffusion models. Furthermore, general generative models lack an understanding of the topological and geometric structures of ground objects in remote sensing imagery. In addition, most diffusion models optimize loss functions at the noise level, leading to a non-intuitive convergence behavior and suboptimal generation quality for complex data. To address these challenges, we propose a Geometric Enhanced Wavelet-based Diffusion Model (GEWDiff), a novel framework for reconstructing hyperspectral images at 4-times super-resolution. A wavelet-based encoder-decoder is introduced that efficiently compresses HSIs into a latent space while preserving spectral-spatial information. To avoid distortion during generation, we incorporate a geometry-enhanced diffusion process that preserves the geometric features. Furthermore, a multi-level loss function was designed to guide the diffusion process, promoting stable convergence and improved reconstruction fidelity. Our model demonstrated state-of-the-art results across multiple dimensions, including fidelity, spectral accuracy, visual realism, and clarity.