# R2H-Diff: Guided Spectral Diffusion Model for RGB-to-Hyperspectral Reconstruction

arXiv: https://arxiv.org/abs/2605.05688 · HTML: https://mlsanity.com/paper/2605.05688/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Inverse problems and image restoration, one of 14 directions in Diffusion models.
- Builds on 25 of its 49 references that are in the map.
- First posted on arXiv 7 May 2026.

## Placed in

- Diffusion models / [Inverse problems and image restoration](https://mlsanity.com/diffusion/inverse-problems-and-image-restoration/)

## Builds on (most cited first, 25 of 25)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020)
- [Generative Adversarial Networks](https://mlsanity.com/paper/2203.00667/) (2022)
- [Scaling Rectified Flow Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2403.03206/) (2024)
- [Restormer: Efficient Transformer for High-Resolution Image Restoration](https://mlsanity.com/paper/2111.09881/) (2021)
- [Enhanced Deep Residual Networks for Single Image Super-Resolution](https://mlsanity.com/paper/1707.02921/) (2017)
- [Image Super-Resolution via Iterative Refinement](https://mlsanity.com/paper/2104.07636/) (2021)
- [Diffusion Models in Vision: A Survey](https://mlsanity.com/paper/2209.04747/) (2022)
- [Simple diffusion: End-to-end diffusion for high resolution images](https://mlsanity.com/paper/2301.11093/) (2023)
- [Learning Enriched Features for Real Image Restoration and Enhancement](https://mlsanity.com/paper/2003.06792/) (2020)
- [Diffusion Model-Based Image Editing: A Survey](https://mlsanity.com/paper/2402.17525/) (2024)
- [Diff-Retinex: Rethinking Low-light Image Enhancement with A Generative Diffusion Model](https://mlsanity.com/paper/2308.13164/) (2023)
- [Diffusion Models in Low-Level Vision: A Survey](https://mlsanity.com/paper/2406.11138/) (2024)
- [MST++: Multi-stage Spectral-wise Transformer for Efficient Spectral Reconstruction](https://mlsanity.com/paper/2204.07908/) (2022)
- [HDNet: High-resolution Dual-domain Learning for Spectral Compressive Imaging](https://mlsanity.com/paper/2203.02149/) (2022)
- [Diff-Plugin: Revitalizing Details for Diffusion-based Low-level Tasks](https://mlsanity.com/paper/2403.00644/) (2024)
- [Self-supervised Neural Networks for Spectral Snapshot Compressive Imaging](https://mlsanity.com/paper/2108.12654/) (2021)
- [Spatial-Spectral Transformer for Hyperspectral Image Denoising](https://mlsanity.com/paper/2211.14090/) (2022)
- [DDS2M: Self-Supervised Denoising Diffusion Spatio-Spectral Model for Hyperspectral Image Restoration](https://mlsanity.com/paper/2303.06682/) (2023)
- [ESSAformer: Efficient Transformer for Hyperspectral Image Super-resolution](https://mlsanity.com/paper/2307.14010/) (2023)
- [Hyperspectral Image Reconstruction via Combinatorial Embedding of Cross-Channel Spatio-Spectral Clues](https://mlsanity.com/paper/2312.11119/) (2023)
- [In2SET: Intra-Inter Similarity Exploiting Transformer for Dual-Camera Compressive Hyperspectral Imaging](https://mlsanity.com/paper/2312.13319/) (2023)
- [Hyperspectral Image Super-Resolution via Dual-domain Network Based on Hybrid Convolution](https://mlsanity.com/paper/2304.04589/) (2023)
- [IT-OSE: Exploring Optimal Sample Size for Industrial Data Augmentation](https://mlsanity.com/paper/2602.15878/) (2026)
- [DS-Diffusion: Data Style-Guided Diffusion Model for Time-Series Generation](https://mlsanity.com/paper/2509.18584/) (2025)

## Abstract (arXiv)

RGB-to-hyperspectral image reconstruction is a highly ill-posed inverse problem, since multiple plausible spectral distributions may correspond to the same RGB observation. Existing regression-based methods usually learn a deterministic mapping, which limits their ability to model reconstruction uncertainty and often leads to over-smoothed spectral responses. Although diffusion models provide strong distribution modeling capability, their direct application to hyperspectral reconstruction remains challenging due to the high spectral dimensionality, strong inter-band correlations, and strict requirement for spectral fidelity. To this end, we propose R2H-Diff, an efficient diffusion-based framework tailored for RGB-to-HSI reconstruction. Specifically, R2H-Diff formulates spectral recovery as a conditional iterative refinement process, enabling progressive reconstruction under RGB guidance. We proposed a Guided Spectral Refinement Module for RGB-conditioned feature fusion and a Hyperspectral-Adaptive Transposed Attention module for efficient spatial--spectral dependency modeling. Furthermore, a normalization-free denoising backbone is adopted to preserve spectral amplitude consistency, while a task-adapted linear noise schedule enables high-quality reconstruction with only five denoising steps. Extensive experiments on NTIRE2022, CAVE, and Harvard demonstrate that R2H-Diff achieves a favorable balance between reconstruction quality and computational efficiency. Notably, on NTIRE2022, R2H-Diff obtains 35.37 dB PSNR with a sub-million-parameter model of 0.58M parameters and 12.25G FLOPs, achieving the lowest model complexity among the evaluated methods while maintaining strong reconstruction fidelity.