Saved in:
Bibliographic Details
Main Authors: Thirgood, Christopher, Mendez, Oscar, Ling, Erin, Storey, Jon, Hadfield, Simon
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2510.16664
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915563008163840
author Thirgood, Christopher
Mendez, Oscar
Ling, Erin
Storey, Jon
Hadfield, Simon
author_facet Thirgood, Christopher
Mendez, Oscar
Ling, Erin
Storey, Jon
Hadfield, Simon
contents Hyperspectral images (HSI) promise to support a range of new applications in computer vision. Recent research has explored the feasibility of generalizable Spectral Reconstruction (SR), the problem of recovering a HSI from a natural three-channel color image in unseen scenarios. However, previous Multi-Scale Attention (MSA) works have only demonstrated sufficient generalizable results for very sparse spectra, while modern HSI sensors contain hundreds of channels. This paper introduces a novel approach to spectral reconstruction via our HYbrid knowledge Distillation and spectral Reconstruction Architecture (HYDRA). Using a Teacher model that encapsulates latent hyperspectral image data and a Student model that learns mappings from natural images to the Teacher's encoded domain, alongside a novel training method, we achieve high-quality spectral reconstruction. This addresses key limitations of prior SR models, providing SOTA performance across all metrics, including an 18\% boost in accuracy, and faster inference times than current SOTA models at various channel depths.
format Preprint
id arxiv_https___arxiv_org_abs_2510_16664
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle HYDRA: HYbrid knowledge Distillation and spectral Reconstruction Algorithm for high channel hyperspectral camera applications
Thirgood, Christopher
Mendez, Oscar
Ling, Erin
Storey, Jon
Hadfield, Simon
Computer Vision and Pattern Recognition
Hyperspectral images (HSI) promise to support a range of new applications in computer vision. Recent research has explored the feasibility of generalizable Spectral Reconstruction (SR), the problem of recovering a HSI from a natural three-channel color image in unseen scenarios. However, previous Multi-Scale Attention (MSA) works have only demonstrated sufficient generalizable results for very sparse spectra, while modern HSI sensors contain hundreds of channels. This paper introduces a novel approach to spectral reconstruction via our HYbrid knowledge Distillation and spectral Reconstruction Architecture (HYDRA). Using a Teacher model that encapsulates latent hyperspectral image data and a Student model that learns mappings from natural images to the Teacher's encoded domain, alongside a novel training method, we achieve high-quality spectral reconstruction. This addresses key limitations of prior SR models, providing SOTA performance across all metrics, including an 18\% boost in accuracy, and faster inference times than current SOTA models at various channel depths.
title HYDRA: HYbrid knowledge Distillation and spectral Reconstruction Algorithm for high channel hyperspectral camera applications
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2510.16664