Towards Realistic Low-Light Image Enhancement via ISP-Driven Data Modeling.

Deep neural networks (DNNs) have recently become the leading method for low-light image enhancement (LLIE). However, despite significant progress, their outputs may still exhibit issues such as amplified noise, incorrect white balance, or unnatural enhancements when deployed in real-world applications. A key challenge is the lack of diverse, large-scale training data that captures the complexities of low-light conditions and imaging pipelines. In this paper, we propose an image signal processing (ISP)-driven data synthesis pipeline that addresses these challenges by generating scalable paired training data. Specifically, our pipeline begins with easily collected high-quality normal-light images, which are first unprocessed into a pseudo-RAW representation using a reverse ISP. We then synthesize low-light degradations directly in this RAW-like domain. The resulting data are subsequently processed through a series of ISP stages, including white balance adjustment, color space conversion, tone mapping, and gamma correction, with controlled variations introduced at each stage. This broadens the degradation space and enhances the diversity of the training data. To demonstrate the effectiveness of the synthetic data, we conduct extensive experiments using a vanilla U-Net model consisting solely of convolutional layers, group normalization, GeLU activation, and convolutional block attention modules (CBAMs). Extensive testing across multiple datasets reveals that the vanilla U-Net model trained with our data synthesis pipeline delivers high-fidelity, visually appealing enhancement results. Additionally, we apply our pipeline to existing LLIE methods, demonstrating its impact on improving their practicality and generalizability in real-world applications. Our codes are publicly available at https://github.com/SMBU-MM/LLIE.