Parallel parsing of MPEG video in heterogeneous distributed environment

Yunyoung Nam, Een Jun Hwang

Research output: Contribution to journalArticle

Abstract

As the use of digital video is getting popular, there is an increasing demand for efficient retrieval of video. To do that, effective video indexing should be incorporated. One of the most fundamental steps in video indexing is decompose video stream into shots and scenes by parsing. Generally, it takes long time to parse video by traditional sequential computers due to the huge amount of computation. In order to solve this problem and speed up the process, we propose three different parallel scheduling algorithms for heterogeneous distributed multicomputer and compare their performance using extensive simulations. We report some of the results in terms of speedup and load balancing.

Original languageEnglish
Pages (from-to)264-274
Number of pages11
JournalLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume2720
Publication statusPublished - 2003 Dec 1
Externally publishedYes

    Fingerprint

ASJC Scopus subject areas

  • Biochemistry, Genetics and Molecular Biology(all)
  • Computer Science(all)
  • Theoretical Computer Science

Cite this