An efficient region based intra prediction algorithm for H.264/advaneed video coding

K. Bharanitharan, Bin-Da Liu, Jar-Ferr Yang

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

H.264, MPEG-4 Part 10, is the latest digital video coding standard that is noted for achieving very high data compression by using several new coding features. One of the key features is the intra prediction based on the rate distortion optimization (RDO) procedure, which requires a true encoding of the intra block by trying all intra prediction modes. The computational burden of intra prediction with the RDO procedure is extremely high for real applications. In this paper, we propose an efficient fast mode decision method by using bi-region similarity detection (BSD) algorithm to reduce the encoding time. The proposed BSD algorithm effectively estimates the texture direction of the block to narrow down the predictive modes to reduce RDO computation. Experimental results show that the proposed algorithm achieves more than a 60% time reduction with negligible PSNR loss and bit-rate increment.

Original languageEnglish
Title of host publicationProceedings of the 2008 International Conference on Computer and Electrical Engineering, ICCEE 2008
Pages393-397
Number of pages5
DOIs
Publication statusPublished - 2008 Dec 1
Event2008 International Conference on Computer and Electrical Engineering, ICCEE 2008 - Phuket, Thailand
Duration: 2008 Dec 202008 Dec 22

Publication series

NameProceedings of the 2008 International Conference on Computer and Electrical Engineering, ICCEE 2008

Other

Other2008 International Conference on Computer and Electrical Engineering, ICCEE 2008
Country/TerritoryThailand
CityPhuket
Period08-12-2008-12-22

All Science Journal Classification (ASJC) codes

  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'An efficient region based intra prediction algorithm for H.264/advaneed video coding'. Together they form a unique fingerprint.

Cite this