Semi-supervised atmospheric component learning in low-light image problem

TL;DR

This paper introduces a semi-supervised method for low-light image enhancement that leverages physical atmospheric models and no-reference quality metrics, achieving state-of-the-art results with improved data efficiency.

Contribution

It proposes a novel semi-supervised training approach that incorporates physical atmospheric models and no-reference metrics for low-light image restoration.

Findings

Achieves state-of-the-art or comparable performance on six datasets.

Utilizes a single objective function for atmospheric component learning.

Demonstrates improved data efficiency over existing methods.

Abstract

Ambient lighting conditions play a crucial role in determining the perceptual quality of images from photographic devices. In general, inadequate transmission light and undesired atmospheric conditions jointly degrade the image quality. If we know the desired ambient factors associated with the given low-light image, we can recover the enhanced image easily \cite{b1}. Typical deep networks perform enhancement mappings without investigating the light distribution and color formulation properties. This leads to a lack of image instance-adaptive performance in practice. On the other hand, physical model-driven schemes suffer from the need for inherent decompositions and multiple objective minimizations. Moreover, the above approaches are rarely data efficient or free of postprediction tuning. Influenced by the above issues, this study presents a semisupervised training method using no-reference image quality metrics for low-light image restoration. We incorporate the classical haze distribution model \cite{b2} to explore the physical properties of the given image in order to learn the effect of atmospheric components and minimize a single objective for restoration. We validate the performance of our network for six widely used low-light datasets. The experiments show that the proposed study achieves state-of-the-art or comparable performance.