A framework for end-to-end video quality prediction of MPEG video

Harilaos Koumaras, C. H. Lin, C. K. Shieh, Anastasios Kourtis

Research output: Contribution to journalArticlepeer-review

18 Citations (Scopus)

Abstract

This paper proposes, describes and evaluates a novel framework for video quality prediction of MPEG-based video services, considering the perceptual degradation that is introduced by the encoding process and the provision of the encoded signal over an error-prone wireless or wire-line network. The concept of video quality prediction is considered in this work, according to which the encoding parameters of the video service and the network QoS conditions are used for performing an estimation/prediction of the video quality level at the user side, without further processing of the actual encoded and transmitted video content. The proposed prediction framework consists of two discrete models: (i) a model for predicting the video quality of an encoded signal at a pre-encoding stage by correlating the spatiotemporal content dynamics to the bit rate that satisfies a specific level of user satisfaction; and (ii) a model that predicts primarily the undecodable frames (and subsequently the perceived quality degradation caused by them) based on the monitored averaged packet loss ratio of the network. The proposed framework is experimentally tested and validated with video signals encoded according to MPEG-4 standard.

Original languageEnglish
Pages (from-to)139-154
Number of pages16
JournalJournal of Visual Communication and Image Representation
Volume21
Issue number2
DOIs
Publication statusPublished - 2010 Feb

All Science Journal Classification (ASJC) codes

  • Signal Processing
  • Media Technology
  • Computer Vision and Pattern Recognition
  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'A framework for end-to-end video quality prediction of MPEG video'. Together they form a unique fingerprint.

Cite this