Decoupled Low-Light Image Enhancement

The visual quality of photographs taken under imperfect lightness conditions can be degenerated by multiple factors, e.g., low lightness, imaging noise, color distortion, and so on. Current low-light image enhancement models focus on the improvement only, or simply deal with all degeneration factors as a whole, therefore leading to sub-optimal results. In this article, we propose decouple model into two sequential stages. first stage focuses improving scene visibility based pixel-wise non-linear mapping. second appearance fidelity suppressing rest factors. decoupled facilitates in aspects. On one hand, whole divided easier subtasks. only aims enhance visibility. It also helps bridge large intensity gap between normal-light images. way, subtask described local adjustment. other since parameter matrix learned from is aware distribution structure, it incorporated complementary information. experiments, our demonstrates state-of-the-art performance both qualitative quantitative comparisons, compared models. addition, ablation studies validate effectiveness aspects, such structure loss function.