DiffLLE: Diffusion-guided Domain Calibration for Unsupervised Low-light Image Enhancement

TL;DR

DiffLLE introduces a diffusion-guided domain calibration framework that enhances unsupervised low-light image enhancement by bridging domain gaps and utilizing priors from pre-trained diffusion models, leading to superior results.

Contribution

The paper proposes a novel diffusion-guided domain calibration method with two plug-and-play modules, improving unsupervised low-light enhancement's robustness and effectiveness.

Findings

Outperforms some supervised methods with only unsupervised baseline.

Effectively narrows the domain gap using diffusion-based calibration.

Achieves superior enhancement quality in diverse real-world scenarios.

Abstract

Existing unsupervised low-light image enhancement methods lack enough effectiveness and generalization in practical applications. We suppose this is because of the absence of explicit supervision and the inherent gap between real-world scenarios and the training data domain. In this paper, we develop Diffusion-based domain calibration to realize more robust and effective unsupervised Low-Light Enhancement, called DiffLLE. Since the diffusion model performs impressive denoising capability and has been trained on massive clean images, we adopt it to bridge the gap between the real low-light domain and training degradation domain, while providing efficient priors of real-world content for unsupervised models. Specifically, we adopt a naive unsupervised enhancement algorithm to realize preliminary restoration and design two zero-shot plug-and-play modules based on diffusion model to improve generalization and effectiveness. The Diffusion-guided Degradation Calibration (DDC) module narrows the gap between real-world and training low-light degradation through diffusion-based domain calibration and a lightness enhancement curve, which makes the enhancement model perform robustly even in sophisticated wild degradation. Due to the limited enhancement effect of the unsupervised model, we further develop the Fine-grained Target domain Distillation (FTD) module to find a more visual-friendly solution space. It exploits the priors of the pre-trained diffusion model to generate pseudo-references, which shrinks the preliminary restored results from a coarse normal-light domain to a finer high-quality clean field, addressing the lack of strong explicit supervision for unsupervised methods. Benefiting from these, our approach even outperforms some supervised methods by using only a simple unsupervised baseline. Extensive experiments demonstrate the superior effectiveness of the proposed DiffLLE.