Abstract
Low-light images often suffer from low contrast, noise, and color distortion, degrading visual quality and impairing downstream vision tasks. We propose a novel conditional diffusion framework for low-light image enhancement that incorporates a Structured Control Embedding Module (SCEM). SCEM decomposes a low-light image into four informative components including illumination, illumination-invariant features, shadow priors, and color-invariant cues. These components serve as control signals that condition a U-Net-based diffusion model trained with a simplified noise-prediction loss. Thus, the proposed SCEM equipped Diffusion method enforces structured enhancement guided by physical priors. In experiments, our model is trained only on the LOLv1 dataset and evaluated without fine-tuning on LOLv2-real, LSRW, DICM, MEF, and LIME. The method achieves state-of-the-art performance in quantitative and perceptual metrics, demonstrating strong generalization across benchmarks. this https URL.
Abstract (translated)
低光图像常常会因为对比度低、噪声多和色彩失真等问题而降低视觉质量，影响下游的视觉任务。我们提出了一种新的条件扩散框架来增强低光图像，并在此框架中引入了一个结构化控制嵌入模块（SCEM）。SCEM 将一幅低光图像分解成四个信息量丰富的组成部分：照明、非光照不变特征、阴影先验和色彩不变线索。这些部分作为控制信号，用于调节一个基于 U-Net 的扩散模型的训练过程，该模型使用简化的噪声预测损失进行训练。因此，所提出的 SCEM 配备的方法通过物理先验指导来执行结构化增强。 在实验中，我们的模型仅在 LOLv1 数据集上进行了训练，并且在没有微调的情况下对 LOLv2-real、LSRW、DICM、MEF 和 LIME 进行了评估。该方法在定量和感知指标上都取得了最先进的性能，在不同基准测试中的泛化能力表现出色。 原文链接：[请在此处插入实际的URL链接，上述描述并未提供具体的链接地址]
URL
https://arxiv.org/abs/2603.00337
