WREN: Low Light Image Enhancement Using Retinex theory-based Double U-Net-like Structures

2026-07-16

AI Summary

This work addresses the challenges of decomposition instability and over-enhancement in low-light image enhancement under complex dynamic range scenarios. Building upon Retinex theory, the authors propose a dual-branch U-Net-like architecture: the first branch decomposes the input image into reflectance and illumination components, while the second branch enhances only the illumination map. A customized Transformer module is introduced to effectively model long-range illumination dependencies. The method rigorously adheres to Retinex assumptions and incorporates a scale-invariant loss function to improve robustness against illumination scaling. Extensive experiments demonstrate that the proposed model achieves state-of-the-art performance across multiple benchmark datasets, significantly enhancing the stability and generalization capability of low-light image restoration.

Abstract

This paper proposes a neural network for low light image enhancement (LLIE) based on retinex theory to make LLIE robust for various dynamic range scenes. The retinex theory is an image formulation model inspired by a human color perception hypothesis, where a low light image is decomposed into intrinsic color context (i.e., reflectance map) and scene-dependent illumination (i.e., illumination map). Due to non-uniqueness of its decomposition, existing retinex-based LLIE methods often fail to achieve stable decomposition, which lead to over-enhancement. Typically, they are sensitive to the dynamic ranges that vary in different lighting conditions. To tackle this issue, we propose WREN: An LLIE neural network with double U-Net-like structures. WREN consists of two U-Net-like sub-networks. The first network has one encoder and two decoders that decompose an input image into the reflectance and illumination maps. The second network with a customized Transformer block between an encoder and a decoder only enhances the illumination map obtained from the first network: This completely follows the assumption of the retinex theory. Finally, the enhanced illumination map is recombined with the reflectance map. The network is trained end-to-end with a scale-invariant loss function, which gives robustness against the illumination scaling. Numerical results show that our method achieves the state-of-the-art performance across multiple datasets. Our code is available online.

Problem

• low light image enhancement
• Retinex theory
• dynamic range
• decomposition instability
• over-enhancement

Innovation

• Retinex theory
• Double U-Net
• Low-light image enhancement
• Illumination enhancement
• Scale-invariant loss