Parallel rate-distortion optimised fast motion estimation algorithm for H.264/AVC using GPU

Muhammad Usman Shahid, Ashfaq Ahmed, Enrico Magli

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

4 Scopus citations

Abstract

Recently, the parallel processing capability of the Graphics Processing Unit (GPU) has been employed for accelerating motion estimation in H.264/AVC encoder implementations. However, while implementing parallel motion estimation on GPU, the bit rate cost of the motion vector is generally ignored. This is due to the unavailability of the spatially predicted motion vectors, which leads to rate-distortion performance degradation. This paper presents a fast parallel motion estimation algorithm implemented on GPU using OpenCL to tackle this problem. The predicted motion vectors are estimated from temporally predicted motion vectors and used for evaluating the bit rate cost of the motion vectors simultaneously. The experimental results show that the proposed scheme achieves significant speedup and has comparable rate-distortion performance with respect to sequential fast motion estimation algorithm.

Original languageBritish English
Title of host publication2013 Picture Coding Symposium, PCS 2013 - Proceedings
PublisherIEEE Computer Society
Pages221-224
Number of pages4
ISBN (Print)9781479902941
DOIs
StatePublished - 2013
Event2013 Picture Coding Symposium, PCS 2013 - San Jose, CA, United States
Duration: 8 Dec 201311 Dec 2013

Publication series

Name2013 Picture Coding Symposium, PCS 2013 - Proceedings

Conference

Conference2013 Picture Coding Symposium, PCS 2013
Country/TerritoryUnited States
CitySan Jose, CA
Period8/12/1311/12/13

Keywords

  • GPU
  • H.264/AVC
  • Motion estimation
  • OpenCL
  • Video coding

Fingerprint

Dive into the research topics of 'Parallel rate-distortion optimised fast motion estimation algorithm for H.264/AVC using GPU'. Together they form a unique fingerprint.

Cite this