Presentation 2007-08-23
Fast Motion Estimation Algorithm and a Low Power Motion Estimation Processor Implementing Dynamic Voltage and Frequency Scaling
Nobuaki Kobayashi, Tadayoshi Enomoto,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) To fully use an advantage of the dynamic voltage and frequency scaling (DVFS), a fast motion estimation (ME) algorithm called adaptively assigned breaking-off condition search (A^2BCS), and an ME processor that employs both A^2BCS and DVFS were developed. The 90-nm CMOS ME processor consists of an absolute difference accumulator and a small DC/DC converter. Power dissipation of the ME processor was 126.9~161.9μW when ME processes at 384kbps, 15fps, p of 10 pixels were carried out using A^2BCS for various CIF test video pictures. It was about 13.36~17.04% that of the ME processor adopting a well known full search ME algorithm.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) motion estimation / DVFS / power dissipation
Paper # SDM2007-144,ICD2007-72
Date of Issue

Conference Information
Committee SDM
Conference Date 2007/8/16(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Silicon Device and Materials (SDM)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Fast Motion Estimation Algorithm and a Low Power Motion Estimation Processor Implementing Dynamic Voltage and Frequency Scaling
Sub Title (in English)
Keyword(1) motion estimation
Keyword(2) DVFS
Keyword(3) power dissipation
1st Author's Name Nobuaki Kobayashi
1st Author's Affiliation Faculty of Science and Engineering, Chuo University()
2nd Author's Name Tadayoshi Enomoto
2nd Author's Affiliation Faculty of Science and Engineering, Chuo University
Date 2007-08-23
Paper # SDM2007-144,ICD2007-72
Volume (vol) vol.107
Number (no) 194
Page pp.pp.-
#Pages 5
Date of Issue