A Switched View of Retinex: Deep Self-Regularized Low-Light Image Enhancement
This work provides a more robust and generalizable low-light image enhancement solution for computer vision applications, particularly in varied lighting conditions.
This paper introduces a self-regularized low-light image enhancement method that addresses color deviation and generalization issues in existing approaches. It achieves this by integrating Retinex theory specifically into the brightness component of an image, while preserving hue and saturation, and demonstrates superior performance over state-of-the-art algorithms.
Self-regularized low-light image enhancement does not require any normal-light image in training, thereby freeing from the chains on paired or unpaired low-/normal-images. However, existing methods suffer color deviation and fail to generalize to various lighting conditions. This paper presents a novel self-regularized method based on Retinex, which, inspired by HSV, preserves all colors (Hue, Saturation) and only integrates Retinex theory into brightness (Value). We build a reflectance estimation network by restricting the consistency of reflectances embedded in both the original and a novel random disturbed form of the brightness of the same scene. The generated reflectance, which is assumed to be irrelevant of illumination by Retinex, is treated as enhanced brightness. Our method is efficient as a low-light image is decoupled into two subspaces, color and brightness, for better preservation and enhancement. Extensive experiments demonstrate that our method outperforms multiple state-of-the-art algorithms qualitatively and quantitatively and adapts to more lighting conditions.