Empowering Low-Light Image Enhancer through Customized Learnable Priors

TL;DR

This paper introduces a novel low-light image enhancement method using customized, learnable priors based on Masked Autoencoders, enhancing transparency, interpretability, and performance over existing approaches.

Contribution

It proposes a new paradigm leveraging MAE-based illumination and noise priors in a deep unfolding framework, improving interpretability and effectiveness.

Findings

Outperforms state-of-the-art methods on multiple datasets

Enhances model interpretability and representation capability

Demonstrates superior noise suppression and brightness enhancement

Abstract

Deep neural networks have achieved remarkable progress in enhancing low-light images by improving their brightness and eliminating noise. However, most existing methods construct end-to-end mapping networks heuristically, neglecting the intrinsic prior of image enhancement task and lacking transparency and interpretability. Although some unfolding solutions have been proposed to relieve these issues, they rely on proximal operator networks that deliver ambiguous and implicit priors. In this work, we propose a paradigm for low-light image enhancement that explores the potential of customized learnable priors to improve the transparency of the deep unfolding paradigm. Motivated by the powerful feature representation capability of Masked Autoencoder (MAE), we customize MAE-based illumination and noise priors and redevelop them from two perspectives: 1) \textbf{structure flow}: we train the MAE from a normal-light image to its illumination properties and then embed it into the proximal operator design of the unfolding architecture; and m2) \textbf{optimization flow}: we train MAE from a normal-light image to its gradient representation and then employ it as a regularization term to constrain noise in the model output. These designs improve the interpretability and representation capability of the model.Extensive experiments on multiple low-light image enhancement datasets demonstrate the superiority of our proposed paradigm over state-of-the-art methods. Code is available at https://github.com/zheng980629/CUE.