# TS-Diff: Two-Stage Diffusion Model for Low-Light RAW Image Enhancement

arXiv: https://arxiv.org/abs/2505.04281 · HTML: https://mlsanity.com/paper/2505.04281/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Inverse problems and image restoration, one of 14 directions in Diffusion models.
- Builds on 39 of its 46 references that are in the map.
- First posted on arXiv 7 May 2025.

## Placed in

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

## Builds on (most cited first, 30 of 39)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [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)
- [Deep Unsupervised Learning using Nonequilibrium Thermodynamics](https://mlsanity.com/paper/1503.03585/) (2015)
- [Image Super-Resolution via Iterative Refinement](https://mlsanity.com/paper/2104.07636/) (2021)
- [RePaint: Inpainting using Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2201.09865/) (2022)
- [Palette: Image-to-Image Diffusion Models](https://mlsanity.com/paper/2111.05826/) (2021)
- [Denoising Diffusion Restoration Models](https://mlsanity.com/paper/2201.11793/) (2022)
- [Noise2Noise: Learning Image Restoration without Clean Data](https://mlsanity.com/paper/1803.04189/) (2018)
- [RepVGG: Making VGG-style ConvNets Great Again](https://mlsanity.com/paper/2101.03697/) (2021)
- [Exploiting Diffusion Prior for Real-World Image Super-Resolution](https://mlsanity.com/paper/2305.07015/) (2023)
- [Learning to See in the Dark](https://mlsanity.com/paper/1805.01934/) (2018)
- [SWAD: Domain Generalization by Seeking Flat Minima](https://mlsanity.com/paper/2102.08604/) (2021)
- [Perception Prioritized Training of Diffusion Models](https://mlsanity.com/paper/2204.00227/) (2022)
- [Toward Convolutional Blind Denoising of Real Photographs](https://mlsanity.com/paper/1807.04686/) (2018)
- [Retinexformer: One-stage Retinex-based Transformer for Low-light Image Enhancement](https://mlsanity.com/paper/2303.06705/) (2023)
- [ACNet: Strengthening the Kernel Skeletons for Powerful CNN via Asymmetric Convolution Blocks](https://mlsanity.com/paper/1908.03930/) (2019)
- [Ultra-High-Definition Low-Light Image Enhancement: A Benchmark and Transformer-Based Method](https://mlsanity.com/paper/2212.11548/) (2022)
- [Learning Enriched Features for Fast Image Restoration and Enhancement](https://mlsanity.com/paper/2205.01649/) (2022)
- [Low-Light Image Enhancement with Wavelet-based Diffusion Models](https://mlsanity.com/paper/2306.00306/) (2023)
- [Diff-Retinex: Rethinking Low-light Image Enhancement with A Generative Diffusion Model](https://mlsanity.com/paper/2308.13164/) (2023)
- [ShadowDiffusion: When Degradation Prior Meets Diffusion Model for Shadow Removal](https://mlsanity.com/paper/2212.04711/) (2022)
- [Global Structure-Aware Diffusion Process for Low-Light Image Enhancement](https://mlsanity.com/paper/2310.17577/) (2023)
- [Low-Light Image Enhancement via Structure Modeling and Guidance](https://mlsanity.com/paper/2305.05839/) (2023)
- [Physics-based Noise Modeling for Extreme Low-light Photography](https://mlsanity.com/paper/2108.02158/) (2021)
- [Pyramid Diffusion Models For Low-light Image Enhancement](https://mlsanity.com/paper/2305.10028/) (2023)
- [Rethinking Noise Synthesis and Modeling in Raw Denoising](https://mlsanity.com/paper/2110.04756/) (2021)
- [Handheld Mobile Photography in Very Low Light](https://mlsanity.com/paper/1910.11336/) (2019)
- [Deep Equilibrium Approaches to Diffusion Models](https://mlsanity.com/paper/2210.12867/) (2022)
- [Abandoning the Bayer-Filter to See in the Dark](https://mlsanity.com/paper/2203.04042/) (2022)

## Built on by (most cited first, 2 of 2)

- [Physics-Guided Rectified Flow for Low-light RAW Image Enhancement](https://mlsanity.com/paper/2509.08330/) (2025)
- [Self-DACE++: Robust Low-Light Enhancement via Efficient Adaptive Curve Estimation](https://mlsanity.com/paper/2604.25367/) (2026)

## Abstract (arXiv)

This paper presents a novel Two-Stage Diffusion Model (TS-Diff) for enhancing extremely low-light RAW images. In the pre-training stage, TS-Diff synthesizes noisy images by constructing multiple virtual cameras based on a noise space. Camera Feature Integration (CFI) modules are then designed to enable the model to learn generalizable features across diverse virtual cameras. During the aligning stage, CFIs are averaged to create a target-specific CFIT, which is fine-tuned using a small amount of real RAW data to adapt to the noise characteristics of specific cameras. A structural reparameterization technique further simplifies CFIT for efficient deployment. To address color shifts during the diffusion process, a color corrector is introduced to ensure color consistency by dynamically adjusting global color distributions. Additionally, a novel dataset, QID, is constructed, featuring quantifiable illumination levels and a wide dynamic range, providing a comprehensive benchmark for training and evaluation under extreme low-light conditions. Experimental results demonstrate that TS-Diff achieves state-of-the-art performance on multiple datasets, including QID, SID, and ELD, excelling in denoising, generalization, and color consistency across various cameras and illumination levels. These findings highlight the robustness and versatility of TS-Diff, making it a practical solution for low-light imaging applications. Source codes and models are available at https://github.com/CircccleK/TS-Diff