LightenDiffusion: Unsupervised Low-Light Image Enhancement with Latent-Retinex Diffusion Models

TL;DR

LightenDiffusion introduces an unsupervised diffusion framework that leverages Retinex theory in latent space for effective low-light image enhancement, outperforming existing methods and achieving results comparable to supervised approaches.

Contribution

The paper presents a novel latent-space Retinex decomposition network integrated with diffusion models for unsupervised low-light enhancement, improving generalization and visual quality.

Findings

Outperforms state-of-the-art unsupervised methods.

Comparable to supervised methods in quality.

Demonstrates strong generalization across scenes.

Abstract

In this paper, we propose a diffusion-based unsupervised framework that incorporates physically explainable Retinex theory with diffusion models for low-light image enhancement, named LightenDiffusion. Specifically, we present a content-transfer decomposition network that performs Retinex decomposition within the latent space instead of image space as in previous approaches, enabling the encoded features of unpaired low-light and normal-light images to be decomposed into content-rich reflectance maps and content-free illumination maps. Subsequently, the reflectance map of the low-light image and the illumination map of the normal-light image are taken as input to the diffusion model for unsupervised restoration with the guidance of the low-light feature, where a self-constrained consistency loss is further proposed to eliminate the interference of normal-light content on the restored results to improve overall visual quality. Extensive experiments on publicly available real-world benchmarks show that the proposed LightenDiffusion outperforms state-of-the-art unsupervised competitors and is comparable to supervised methods while being more generalizable to various scenes. Our code is available at https://github.com/JianghaiSCU/LightenDiffusion.