Assessment of 3D quality of experience with content analysis

Bochao Zou, Yue Liu, Yi Huang, Yongtian Wang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Stereoscopic display has witnessed rapid progress in recent years, but it still suffers from the bad quality of experience (QoE) problem. Development of more reliable metrics for objective assessment of 3D QoE is challenging. In this paper, we propose a model that combines 2D and 3D features of the analyzed images with SSIM in order to produce more reliable estimation of the perceived subjective quality in terms of mean opinion score (MOS). The model was trained and evaluated on a public stereoscopic image database with associated ground truth MOS. Our results show that the proposed model achieved high correlation with perceived quality and it is promising for objective evaluation of large stereoscopic image dataset after a relatively small training procedure.

Original languageEnglish
Title of host publication2013 International Conference on Optical Instruments and Technology
Subtitle of host publicationOptical Systems and Modern Optoelectronic Instruments
PublisherSPIE
ISBN (Print)9780819499608
DOIs
Publication statusPublished - 2013
Event2013 International Conference on Optical Instruments and Technology: Optical Systems and Modern Optoelectronic Instruments - Beijing, China
Duration: 17 Nov 201319 Nov 2013

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume9042
ISSN (Print)0277-786X
ISSN (Electronic)1996-756X

Conference

Conference2013 International Conference on Optical Instruments and Technology: Optical Systems and Modern Optoelectronic Instruments
Country/TerritoryChina
CityBeijing
Period17/11/1319/11/13

Keywords

  • MOS
  • quality of experience
  • stereoscopic 3D
  • structural similarity

Fingerprint

Dive into the research topics of 'Assessment of 3D quality of experience with content analysis'. Together they form a unique fingerprint.

Cite this