Encoder adaptable difference detection for low power video compression in surveillance system

Xin Jin*, Satoshi Goto

*この研究の対応する著者

研究成果

抄録

In this paper, a difference detection algorithm is proposed to reduce the computational complexity and power consumption in surveillance video compression. The content differences of the input video data are automatically detected by analyzing the color and moving correlation features. Macroblocks without content differences are directly distributed to the bitstream writer of the H.264/AVC encoder. Both the computational complexity and the power consumption are significantly reduced by skipping the entire encoding process. An average of over 84% of overall encoding complexity can be reduced. No loss is observed in both of subjective and objective video quality. Without any requirement in changing the encoder hardware, the proposed algorithm provides high adaptability to be integrated into the existing H.264/AVC video encoders.

本文言語English
ホスト出版物のタイトルLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
ページ285-296
ページ数12
6298 LNCS
PART 2
DOI
出版ステータスPublished - 2010
イベント11th Pacific Rim Conference on Multimedia, PCM 2010 - Shanghai
継続期間: 2010 9 212010 9 24

出版物シリーズ

名前Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
番号PART 2
6298 LNCS
ISSN(印刷版)03029743
ISSN(電子版)16113349

Other

Other11th Pacific Rim Conference on Multimedia, PCM 2010
CityShanghai
Period10/9/2110/9/24

ASJC Scopus subject areas

  • コンピュータ サイエンス(全般)
  • 理論的コンピュータサイエンス

フィンガープリント

「Encoder adaptable difference detection for low power video compression in surveillance system」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル