Back to articles
Articles
Volume: 27 | Article ID: art00021
Image
Evaluating Colour Constancy on the new MIST dataset of Multi-Illuminant Scenes
Xiangpeng Hao
Brian Funt
Hanxiao Jiang
  DOI :  10.2352/issn.2169-2629.2019.27.21  Published Online :  October 2019
Abstract

A new image test set of synthetically generated, full-spectrum images with pixelwise ground truth has been developed to aid in the evaluation of illumination estimation methods for colour constancy. The performance of 9 illumination methods is reported for this dataset along and compared to the optimal single-illuminant estimate. None of the methods specifically designed to handle multi-illuminant scenes is found to perform any better than the optimal single-illuminant case based on completely uniform illumination.

Journal Title : Color and Imaging Conference
Publisher Name : Society for Imaging Science and Technology
Subject Areas :
Views 18
Downloads 4
 DOI 10.2352/issn.2169-2629.2019.27.21
 articleview.views 18
 articleview.downloads 4
  Cite this article 

Xiangpeng Hao, Brian Funt, Hanxiao Jiang, "Evaluating Colour Constancy on the new MIST dataset of Multi-Illuminant Scenes"  in Proc. IS&T 27th Color and Imaging Conf.,  2019,  pp 108 - 113,  https://doi.org/10.2352/issn.2169-2629.2019.27.21

 Copy citation
  Copyright statement 
Copyright © Society for Imaging Science and Technology 2019
72010350
Color and Imaging Conference
color imaging conf
2166-9635
Society for Imaging Science and Technology
10.2352/issn.2169-2629.2019.27.21
2166-9635(20191021)2019:1L.108;1-
s21.phd
/ist/cic/2019/00002019/00000001/art00021
Articles
Evaluating Colour Constancy on the new MIST dataset of Multi-Illuminant Scenes
HaoXiangpeng
FuntBrian
JiangHanxiao
21102019
2019
1
108
113
2019
A new image test set of synthetically generated, full-spectrum images with pixelwise ground truth has been developed to aid in the evaluation of illumination estimation methods for colour constancy. The performance of 9 illumination methods is reported for this dataset along and compared to the optimal single-illuminant estimate. None of the methods specifically designed to handle multi-illuminant scenes is found to perform any better than the optimal single-illuminant case based on completely uniform illumination.