Motion estimation for H.264/AVC using programmable graphics hardware

Chi Wang Ho*, Oscar C. Au, S. H.Gary Chan, Shu Kei Yip, Hoi Ming Wong

*Corresponding author for this work

Research output: Chapter in Book/Conference Proceeding/ReportConference Paper published in a bookpeer-review

Abstract

We present an efficient implementation of motion estimation (ME) for H.264/AVC using programmable graphics hardware. The cost function for ME in H.264/AVC depends on the motion vector (MV) predictor which is the median MV of three neighboring coded blocks. Previous implementations assume no dependency among adjacent blocks, which is not true for H.264/AVC, they also perform unsatisfactorily because of their low arithmetic intensity, which is defined as operation per word transferred. To overcome the dependency problem, we introduce a new implementation which performs ME on block-by-block basis. Moreover, we can adjust the arithmetic intensity easily to optimize the performance on different graphics cards. Experimental results show that our implementation is substantially faster (by 10 times) than our SIMD optimized CPU implementation.

Original languageEnglish
Title of host publication2006 IEEE International Conference on Multimedia and Expo, ICME 2006 - Proceedings
Pages2049-2052
Number of pages4
DOIs
Publication statusPublished - 2006
Event2006 IEEE International Conference on Multimedia and Expo, ICME 2006 - Toronto, ON, Canada
Duration: 9 Jul 200612 Jul 2006

Publication series

Name2006 IEEE International Conference on Multimedia and Expo, ICME 2006 - Proceedings
Volume2006

Conference

Conference2006 IEEE International Conference on Multimedia and Expo, ICME 2006
Country/TerritoryCanada
CityToronto, ON
Period9/07/0612/07/06

Fingerprint

Dive into the research topics of 'Motion estimation for H.264/AVC using programmable graphics hardware'. Together they form a unique fingerprint.

Cite this