A Lightweight Low-Light Image Enhancement Network via Channel Prior and Gamma Correction

TL;DR

This paper presents CPGA-Net, a lightweight deep learning model for low-light image enhancement that combines traditional priors with gamma correction, achieving superior performance with minimal parameters and fast inference.

Contribution

The paper introduces a novel LLIE network integrating channel priors and gamma correction, along with an efficient knowledge distillation approach for real-time applications.

Findings

CPGA-Net outperforms existing methods on objective and subjective metrics.

The lightweight model achieves high efficiency with only 0.025 million parameters.

Knowledge distillation further reduces parameters to 0.018 million with faster inference.

Abstract

Human vision relies heavily on available ambient light to perceive objects. Low-light scenes pose two distinct challenges: information loss due to insufficient illumination and undesirable brightness shifts. Low-light image enhancement (LLIE) refers to image enhancement technology tailored to handle this scenario. We introduce CPGA-Net, an innovative LLIE network that combines dark/bright channel priors and gamma correction via deep learning and integrates features inspired by the Atmospheric Scattering Model and the Retinex Theory. This approach combines the use of traditional and deep learning methodologies, designed within a simple yet efficient architectural framework that focuses on essential feature extraction. The resulting CPGA-Net is a lightweight network with only 0.025 million parameters and 0.030 seconds for inference time, yet it achieves superior performance over existing LLIE methods on both objective and subjective evaluation criteria. Furthermore, we utilized knowledge distillation with explainable factors and proposed an efficient version that achieves 0.018 million parameters and 0.006 seconds for inference time. The proposed approaches inject new solution ideas into LLIE, providing practical applications in challenging low-light scenarios.