Skip to main navigation Skip to search Skip to main content

Priority-based selective H.264 SVC streaming over erroneous converged networks

  • InterDigital, Inc.
  • University of Pennsylvania

Research output: Contribution to journal › Article › peer-review

Abstract

This paper provides a detailed description and discussion of new optimal H.264 scalable video coding (SVC) transmission method over multi-path networks that have variable packet loss rates (PLR). The proposed method has three steps: (1) using flexible macroblock ordering (FMO), it prioritizes SVC layers and slice groups (SG) according to their affect on video quality; (2) it measures current channel status and predicts future bandwidths (BW); and (3) it allocates SVC layers and SGs to the prioritized channels by PLR. Experiments show that the proposed selective streaming method can improve video quality as much as 3.4 dB in peak signal-to-noise ratio (PSNR), and has better error resilience than traditional streaming methods.

Original languageEnglish
Pages (from-to)337-353
Number of pages17
JournalMultimedia Tools and Applications
Volume68
Issue number2
DOIs
StatePublished - Jan 2014
Externally publishedYes

Keywords

  • Channel prediction
  • Converged network
  • FMO
  • H.264 SVC
  • Video streaming

Fingerprint

Dive into the research topics of 'Priority-based selective H.264 SVC streaming over erroneous converged networks'. Together they form a unique fingerprint.

Cite this