# RGB Pre-Training Enhanced Unobservable Feature Latent Diffusion Model for Spectral Reconstruction

arXiv: https://arxiv.org/abs/2507.12967 · HTML: https://mlsanity.com/paper/2507.12967/

## 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 17 of its 48 references that are in the map.
- First posted on arXiv 17 Jul 2025.

## Placed in

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

## Builds on (most cited first, 17 of 17)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020)
- [Diffusion Models Beat GANs on Image Synthesis](https://mlsanity.com/paper/2105.05233/) (2021)
- [Image Super-Resolution via Iterative Refinement](https://mlsanity.com/paper/2104.07636/) (2021)
- [Multi-Stage Progressive Image Restoration](https://mlsanity.com/paper/2102.02808/) (2021)
- [Repurposing Diffusion-Based Image Generators for Monocular Depth Estimation](https://mlsanity.com/paper/2312.02145/) (2023)
- [Mask-guided Spectral-wise Transformer for Efficient Hyperspectral Image Reconstruction](https://mlsanity.com/paper/2111.07910/) (2021)
- [Diffusion Models for Image Restoration and Enhancement: A Comprehensive Survey](https://mlsanity.com/paper/2308.09388/) (2023)
- [MST++: Multi-stage Spectral-wise Transformer for Efficient Spectral Reconstruction](https://mlsanity.com/paper/2204.07908/) (2022)
- [Prompt-tuning latent diffusion models for inverse problems](https://mlsanity.com/paper/2310.01110/) (2023)
- [HDNet: High-resolution Dual-domain Learning for Spectral Compressive Imaging](https://mlsanity.com/paper/2203.02149/) (2022)
- [HIR-Diff: Unsupervised Hyperspectral Image Restoration Via Improved Diffusion Models](https://mlsanity.com/paper/2402.15865/) (2024)
- [AdaptiveWeighted Attention Network with Camera Spectral Sensitivity Prior for Spectral Reconstruction from RGB Images](https://mlsanity.com/paper/2005.09305/) (2020)
- [A survey on computational spectral reconstruction methods from RGB to hyperspectral imaging](https://mlsanity.com/paper/2106.15944/) (2021)
- [Hyperspectral recovery from RGB images using Gaussian Processes](https://mlsanity.com/paper/1801.04654/) (2018)
- [Pixel-aware Deep Function-mixture Network for Spectral Super-Resolution](https://mlsanity.com/paper/1903.10501/) (2019)

## Abstract (arXiv)

Spectral reconstruction (SR) is a crucial problem in image processing that requires reconstructing hyperspectral images (HSIs) from the corresponding RGB images. A key difficulty in SR is estimating the unobservable feature, which encapsulates significant spectral information not captured by RGB imaging sensors. The solution lies in effectively constructing the spectral-spatial joint distribution conditioned on the RGB image to complement the unobservable feature. Since HSIs share a similar spatial structure with the corresponding RGB images, it is rational to capitalize on the rich spatial knowledge in RGB pre-trained models for spectral-spatial joint distribution learning. To this end, we extend the RGB pre-trained latent diffusion model (RGB-LDM) to an unobservable feature LDM (ULDM) for SR. As the RGB-LDM and its corresponding spatial autoencoder (SpaAE) already excel in spatial knowledge, the ULDM can focus on modeling spectral structure. Moreover, separating the unobservable feature from the HSI reduces the redundant spectral information and empowers the ULDM to learn the joint distribution in a compact latent space. Specifically, we propose a two-stage pipeline consisting of spectral structure representation learning and spectral-spatial joint distribution learning to transform the RGB-LDM into the ULDM. In the first stage, a spectral unobservable feature autoencoder (SpeUAE) is trained to extract and compress the unobservable feature into a 3D manifold aligned with RGB space. In the second stage, the spectral and spatial structures are sequentially encoded by the SpeUAE and the SpaAE, respectively. The ULDM is then acquired to model the distribution of the coded unobservable feature with guidance from the corresponding RGB images. Experimental results on SR and downstream relighting tasks demonstrate that our proposed method achieves state-of-the-art performance.