Low-Light Image Restoration Based on Retina Model using Neural Networks

TL;DR

This paper introduces a neural network model inspired by the retina that efficiently restores low-light images, achieving comparable perceptual quality to complex models with reduced computational cost.

Contribution

The work demonstrates a simple neural network based on retina principles for low-light image restoration, avoiding manual parameter tuning and aligning with neurobiological organization.

Findings

Achieves perceptually comparable results to complex models

Reduces computational overhead in image restoration

Provides a neurobiologically inspired approach

Abstract

We report the possibility of using a simple neural network for effortless restoration of low-light images inspired by the retina model, which mimics the neurophysiological principles and dynamics of various types of optical neurons. The proposed neural network model saves the cost of computational overhead in contrast with traditional signal-processing models, and generates results comparable with complicated deep learning models from the subjective perceptual perspective. This work shows that to directly simulate the functionalities of retinal neurons using neural networks not only avoids the manually seeking for the optimal parameters, but also paves the way to build corresponding artificial versions for certain neurobiological organizations.