Back to articles
Articles
Volume: 27 | Article ID: art00047
Image
Benchmark of 2D quality metrics for the assessment of 360-deg images
Mohamed-Chaker Larabi
Audrey Girard
Sami Jaballah
Fan Yu
  DOI :  10.2352/issn.2169-2629.2019.27.47  Published Online :  October 2019
Abstract

Omnidirectional or 360-degree images are becoming very popular in many applications and several challenges are raised because of both the nature and the representation of the data. Quality assessment is one of them from two different points of view: objectively or subjectively. In this paper, we propose to study the performance of different metrics belonging to various categories including simple mathematical metrics, humand perception based metrics and spherically optimized metrics. The performance of these metrics is measured using different tools such as PLCC, SROCC, KROCC and RMSE based on the only publically available database from Nanjing university. The results show that the metric that are considered as optimized for 360 degrees images are not providing the best correlation with the human judgement of the quality.

Journal Title : Color and Imaging Conference
Publisher Name : Society for Imaging Science and Technology
Subject Areas :
Views 15
Downloads 2
 DOI 10.2352/issn.2169-2629.2019.27.47
 articleview.views 15
 articleview.downloads 2
  Cite this article 

Mohamed-Chaker Larabi, Audrey Girard, Sami Jaballah, Fan Yu, "Benchmark of 2D quality metrics for the assessment of 360-deg images"  in Proc. IS&T 27th Color and Imaging Conf.,  2019,  pp 262 - 267,  https://doi.org/10.2352/issn.2169-2629.2019.27.47

 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.2169-2629.2019.27.47
2166-9635(20191021)2019:1L.262;1-
s47.phd
/ist/cic/2019/00002019/00000001/art00047
Articles
Benchmark of 2D quality metrics for the assessment of 360-deg images
LarabiMohamed-Chaker
GirardAudrey
JaballahSami
YuFan
21102019
2019
1
262
267
2019
Omnidirectional or 360-degree images are becoming very popular in many applications and several challenges are raised because of both the nature and the representation of the data. Quality assessment is one of them from two different points of view: objectively or subjectively. In this paper, we propose to study the performance of different metrics belonging to various categories including simple mathematical metrics, humand perception based metrics and spherically optimized metrics. The performance of these metrics is measured using different tools such as PLCC, SROCC, KROCC and RMSE based on the only publically available database from Nanjing university. The results show that the metric that are considered as optimized for 360 degrees images are not providing the best correlation with the human judgement of the quality.