RDDM: Practicing RAW Domain Diffusion Model for Real-world Image Restoration

The official implementation for "RDDM: Practicing RAW Domain Diffusion Model for Real-world Image Restoration"

<a href="https://arxiv.org/pdf/2508.19154"> <img src="https://img.shields.io/badge/RDDM-Arxiv-red?logo=arxiv&logoColor=red" alt="RDDM Paper on arXiv" /> </a>

RDDM: Practicing RAW Domain Diffusion Model for Real-world Image Restoration <br>

<div> Yan Chen^1,†&emsp; Yi Wen^1†&emsp; Wei Li¹&emsp; Junchao Liu¹&emsp; Yong Guo²&emsp; Jie Hu¹&emsp; Xinghao Chen¹&emsp; </div> <div> ¹Huawei Noah’s Ark Lab, ²Max Planck Institute for Informatics <br/> </div>

---

Abstract: We present the RAW domain diffusion model (RDDM), an end-to-end diffusion model that restores photo-realistic images directly from the sensor RAW data. While recent sRGB-domain diffusion methods achieve impressive results, they are caught in a dilemma between high fidelity and image generation. As these models process lossy sRGB inputs and neglect the accessibility of the sensor RAW images in many scenarios, e.g., in image and video capturing in edge devices, resulting in sub-optimal performance. RDDM obviates this limitation by directly restoring images in the RAW domain, replacing the conventional two-stage image signal processing (ISP)$\rightarrow$Image Restoration (IR) pipeline. However, a simple adaptation of pre-trained diffusion models to the RAW domain confronts many challenges. To this end, we propose: (1) a RAW-domain VAE (RVAE), encoding sensor RAW and decoding it into an enhanced linear domain image, to solve the out-of-distribution (OOD) issues between the different domain distributions; (2) a configurable multi-bayer (CMB) LoRA module, adapting diverse RAW Bayer patterns such as RGGB, BGGR, etc. To compensate for the deficiency in the dataset, we develop a scalable data synthesis pipeline synthesizing RAW LQ-HQ pairs from existing sRGB datasets for large-scale training. Extensive experiments demonstrate RDDM's superiority over state-of-the-art sRGB diffusion methods, yielding higher fidelity results with fewer artifacts.

Overview

News

• [2025.11] This repo is created.

Dependencies & Installation

Please refer to the following simple steps for installation.

git clone https://github.com/YanCHEN-fr/RDDM.git
cd RDDM
conda create -n RDDM python=3.10 -y
conda activate RDDM
pip install -r requirements.txt

Datasets

Training

cd RDDM
bash train.sh

Test

bash test.sh

Results

Qualitative Comparisons on real RAW dataset (DND)

Qualitative Comparisons on real RAW dataset (SIDD)

Qualitative Comparisons on real RAW dataset (RealCapture)

Qualitative Comparisons with diffusion-based methods

Qualitative Comparisons with GAN-based methods

Quantitative Comparisons

Acknowledgement

This work is released under the Apache 2.0 license.