Skip to main navigation Skip to search Skip to main content

Rate-distortion modeling and its application to quality layer assignment in SVC/H. 264 fine-granular snr scalable videos

  • Peking University

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

Abstract

To support efficient quality extraction in H. 264/AVC-based scalable video coding, it's important to obtain the rate-distortion (R-D) or distortion-rate (D-R) function of each FGS packet. In this paper, firstly, we analyze the sub-bitplane technology of SVC FGS coding, and conclude the MSE-based D-R function should be linear within a FGS layer, which also explains the slow-start phenomena of PSNR-based D-R function. Consequently, a piece-wise linear model is proposed to describe the no-drift R-D function of FGS EL. Secondly, considering both the picture prediction structure and the reference picture correlation, we investigate the SVC drift of hierarchical B pictures and propose a simple and effective distortion model to estimate the reconstructed frame distortion with drift. At last, based on above analysis and models, a new model-based quality layer assignment algorithm is proposed to achieve equivalent coding efficiency as the SVC test model with significantly reduced complexity.

Original languageEnglish
Title of host publicationPCS 2007 - 26th Picture Coding Symposium
StatePublished - 2007
Event26th Picture Coding Symposium, PCS 2007 - Lisbon, Portugal
Duration: 7 Nov 20079 Nov 2007

Publication series

NamePCS 2007 - 26th Picture Coding Symposium

Conference

Conference26th Picture Coding Symposium, PCS 2007
Country/TerritoryPortugal
CityLisbon
Period7/11/079/11/07

Keywords

  • Drift analysis
  • FGS
  • H. 264/AVC
  • Quality layer
  • Rate-distortion model
  • SVC

Fingerprint

Dive into the research topics of 'Rate-distortion modeling and its application to quality layer assignment in SVC/H. 264 fine-granular snr scalable videos'. Together they form a unique fingerprint.

Cite this