Introduction

This is the website for our paper "Revisit Retinex Theory: Towards A Lightness-Aware Restorer for Underexposed Images".

Abstract

We investigate how to correct exposure of underexposed images. The bottleneck of previous methods mainly lies in their naturalness and robustness when dealing with images with various exposure levels. When facing well-exposed or extremely underexposed images, they may produce over- or under-enhanced outputs. In this paper, we propose a novel Retinex-based approach, namely LiAR (short for Lightness-Aware Restorer). The word ``lightness-aware'' refers to that the estimated illumination not only is a component to be adjusted, but also serves as a measure that reflects the brightness of the scene, determining the degree of adjustment. In this way, underexposed images can be restored adaptively according to their own brightness. Given an image, LiAR first estimates its illumination map using a specially designed loss function which can ensure the result's color consistency and texture richness. Then adaptive correction is performed to get properly exposed output. LiAR is based on internal optimization of the single test image and doesn't need any prior training, implying that it can adapt itself to different settings per image. Additionally, LiAR can be easily extended to the video case due to its simplicity and stability. Experiments demonstrate that facing images/videos with various exposure levels, LiAR can achieve robust and real-time correction with high contrast and naturalness.

Source Codes and Video Dataset

Get the code and dataset. Extraction code: n120

Dependencies

Python3

PyTorch >= 0.4.1

PIL >= 6.1.0

Opencv-python>=3.4

Last update: Jun. 6, 2020