Skip to main navigation Skip to search Skip to main content

Cross-lab subjective evaluation of the MVC+D and 3D-AVC 3D video coding standards

  • P. Hanhart
  • , N. Ramzan
  • , V. Baroncini
  • , T. Ebrahimi

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Abstract

    Consistent and imitable subjective measurement 3D video quality assessment is investigated for evaluating 3D service parameters and as an essential criterion towards the development of objective models. This paper analyzes the results obtained from three test laboratories to evaluate the performance of the MVC+D and 3D-AVC 3D video coding standards using identical video content and following similar methodologies and instructions. The correlation between laboratories was investigated using similar analysis to benchmarking of objective metrics. Analyses show that test laboratories employing different displays and different subjects could still produce highly correlated results, as they follow similar guidelines to carry out assessments.
    Original languageEnglish
    Title of host publicationSixth International Workshop on Quality of Multimedia Experience (QoMEX), 2014
    PublisherIEEE
    Pages183-188
    Number of pages6
    ISBN (Print)9781479965366
    DOIs
    Publication statusPublished - 2014

    Keywords

    • performance evaluation
    • standards
    • video coding
    • 3D service parameter evaluation
    • 3D video quality assessment
    • 3D-AVC 3D video coding standards
    • MVC+D video coding standards
    • cross-lab subjective evaluation
    • identical video content
    • Bit rate
    • Correlation
    • Laboratories
    • Multimedia communication
    • Quality assessment
    • Three-dimensional displays
    • 3D-AVC
    • MVC+D
    • Subjective evaluation
    • cross-lab validation
    • viewing environment

    Fingerprint

    Dive into the research topics of 'Cross-lab subjective evaluation of the MVC+D and 3D-AVC 3D video coding standards'. Together they form a unique fingerprint.

    Cite this