Multi-illuminant Color Constancy via Multi-scale Illuminant Estimation and Fusion

TL;DR

This paper introduces a multi-scale, multi-branch neural network approach for multi-illuminant color constancy, effectively estimating and fusing illuminant maps across scales to improve local color correction.

Contribution

It proposes a novel tri-branch convolution network with an attentional fusion module for multi-scale illuminant estimation, addressing scale impact neglected by prior methods.

Findings

Achieves state-of-the-art performance on benchmark datasets.

Demonstrates improved accuracy over existing deep learning methods.

Validates effectiveness through comprehensive experimental analysis.

Abstract

Multi-illuminant color constancy methods aim to eliminate local color casts within an image through pixel-wise illuminant estimation. Existing methods mainly employ deep learning to establish a direct mapping between an image and its illumination map, which neglects the impact of image scales. To alleviate this problem, we represent an illuminant map as the linear combination of components estimated from multi-scale images. Furthermore, we propose a tri-branch convolution networks to estimate multi-grained illuminant distribution maps from multi-scale images. These multi-grained illuminant maps are merged adaptively with an attentional illuminant fusion module. Through comprehensive experimental analysis and evaluation, the results demonstrate the effectiveness of our method, and it has achieved state-of-the-art performance.