No-Reference Video Quality Assessment using MPEG Analysis

Jacob Søgaard, Søren Forchhammer, Jari Korhonen

    Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review


    We present a method for No-Reference (NR) Video Quality Assessment (VQA) for decoded video without access to the bitstream. This is achieved by extracting and pooling features from a NR image quality assessment method used frame by frame. We also present methods to identify the video coding and estimate the video coding parameters for MPEG-2 and H.264/AVC which can be used to improve the VQA. The analysis differs from most other video coding analysis methods since it is without access to the bitstream. The results show that our proposed method is competitive with other recent NR VQA methods for MPEG-2 and H.264/AVC.
    Original languageEnglish
    Title of host publication2013 Picture Coding Symposium (PCS)
    Publication date2013
    ISBN (Print)978-1-4799-0292-7
    Publication statusPublished - 2013
    EventPicture Coding Symposium (PCS 2013) - San Jose, CA, United States
    Duration: 8 Dec 201311 Dec 2013


    ConferencePicture Coding Symposium (PCS 2013)
    Country/TerritoryUnited States
    CitySan Jose, CA
    Internet address


    Dive into the research topics of 'No-Reference Video Quality Assessment using MPEG Analysis'. Together they form a unique fingerprint.

    Cite this