DocumentCode :
3196017
Title :
Local Reference with Early Termination in H.264 Motion Estimation
Author :
Xiao Su ; Singh, Sweta ; Bai, Yan
Author_Institution :
San Jose State Univ., San Jose
fYear :
2007
fDate :
2-5 July 2007
Firstpage :
384
Lastpage :
387
Abstract :
Multiple reference frames and variable block sizes improve compression efficiency of H.264, however, they also increase the encoder complexity and motion estimation time. This paper proposes a new algorithm, called local reference with early termination (LRET) to reduce the H.264 motion estimation time without adding to the encoder complexity. The LERT algorithm rearranges the search order of the reference frames based on the selection probability of the reference frames in the current frame. The experimental results show that the LERT achieves up to 59% reduction in motion estimation time with comparable video quality and negligible increase in bit-rate, as compared to the best algorithm in H.264 reference software.
Keywords :
computational complexity; data compression; motion estimation; video coding; H.264 motion estimation; compression efficiency; encoder complexity; local reference with early termination algorithm; multiple reference frames; variable block sizes; video compression; Algorithm design and analysis; Costs; Decoding; Encoding; History; Motion analysis; Motion estimation; Quantization; Statistical analysis; Video compression;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Multimedia and Expo, 2007 IEEE International Conference on
Conference_Location :
Beijing
Print_ISBN :
1-4244-1016-9
Electronic_ISBN :
1-4244-1017-7
Type :
conf
DOI :
10.1109/ICME.2007.4284667
Filename :
4284667
Link To Document :
بازگشت