欢迎访问中国科学院大学学报,今天是
信息与电子科学

细粒度并行归一化部分失真运动估计

  • 袁竞杰 ,
  • 张清毅 ,
  • 马宜科 ,
  • 宋风龙
展开
  • 1. 中国矿业大学(北京)机电与信息工程学院, 北京 100083;
    2. 中国科学院计算技术研究所计算机体系结构国家重点实验室, 北京 100190

收稿日期: 2012-12-19

  修回日期: 2013-03-29

  网络出版日期: 2013-11-15

基金资助

国家自然科学基金(61100013)和中央高校基本科研业务费专项(2010YJ19)资助

Fine-grained parallel algorithm for normalized partial distortion search

  • YUAN Jing-Jie ,
  • ZHANG Qing-Yi ,
  • MA Yi-Ke ,
  • SONG Feng-Long
Expand
  • 1. College of Mechanical and Electrical Engineering, China University of Mining & Technology, Beijing 100083, China;
    2. State Key Laboratory of Computer Architecture, Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190, China

Received date: 2012-12-19

  Revised date: 2013-03-29

  Online published: 2013-11-15

摘要

移动视频编码应用对实时性要求越来越高,传统编码器中使用的串行运动估计算法难以满足实时编码要求.本文并行化移动编码中典型的运动估计算法——归一化部分失真搜索.采用比帧和宏块更小的候选块作为并行粒度,保持归一化部分失真快速排除非最佳候选块优势,同时充分利用多核计算资源.4核CPU平台上实验结果表明,相比串行算法,该并行算法在计算量增加不超过1.2%的前提下,实现了3.88至3.96的加速比.

本文引用格式

袁竞杰 , 张清毅 , 马宜科 , 宋风龙 . 细粒度并行归一化部分失真运动估计[J]. 中国科学院大学学报, 2013 , 30(6) : 813 -818 . DOI: 10.7523/j.issn.2095-6134.2013.06.015

Abstract

Serial motion estimation does not satisfy real-time requirements of video coding in mobile terminal. A divided-spiral-path parallel motion estimation based on normalized partial distortion search is proposed to make full use of multi-core computing power, and it retains low computational complexity of the serial algorithm. Experimental results show that the proposed algorithm achieves speedup ratios of 3.88-3.96 on a 4-core CPU platform while the cost of trivial computational effort increases by less than 1.2%.

参考文献

[1] Schwalb M, Ewerth R, Freisleben B. Fast motion estimation on graphics hardware for H.264 video encoding[J]. IEEE Transactions on multimedia, 2009, 11(1): 1-10.

[2] Chen Y K, Li E Q,Zhou X S,et al. Implementation of H.264 encoder and decoder on personal computers[J]. IEEE Transactions on multimedia, 2006, 17(2): 509-532.

[3] Lin Y C, Li P L, Chang C H, et al. Multi-pass algorithm of motion estimation in video encoding for generic GPU[C]//Proc of IEEE International Symposium on Circuits and Systems (ISCAS). 2006: 1-4.

[4] Cheung N M, Fan X P, Au O C, et al. Video coding on multicore graphics processors[J]. IEEE Signal Processing Magazine, 2010, 27(2): 79-89.

[5] Momcilovic S, Sousa L. A parallel algorithm for advanced video motion estimation on multicore architectures[C]//Proc of International Conference on Complex, Intelligent and Software Intensive Systems (CISIS). 2008: 831-836.

[6] Kung M C, Au O C, Wong P H W, et al. Block based parallel motion estimation using programmable graphics hardware[C]//Proc of International Conference on Audio, Language and Image Processing (ICALIP). 2008: 599-603.

[7] Chen Y B, Li Z D, Guo L, et al. Architecture design of low-power motion estimation based on DHS-NPDS for H.264/AVC[J]. Science China Information Fi Science, 2012, 42(4): 527-536 (in Chinese). 陈运必,李正东,郭立,等. 基于DHS-NPDS 的低功耗运动估计硬件结构设计[J]. 中国科学F辑:信息科学,2012,42(4): 527-536.

[8] Chen T C, Chien S Y, Huang Y W, et al. Analysis and architecture design of an HDTV720p 30 frames/s H.264/AVC encoder[J]. IEEE Trans Circ Syst Vid, 2006(16): 673-688.

[9] Urhan O. Constrained one-bit transform based fast block motion estimation using adaptive search range[J]. IEEE Transactions on Consumer Electronics, 2010, 56(3): 1 868-1 871.

[10] Cheung C K, Po L M. Normalized partial distortion search algorithm for block motion estimation[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2000, 10(3): 417-422.

[11] Yi X Q, Ling N. Improved normalized partial distortion search with dual-halfway-stop for rapid block motion estimation[J]. IEEE Transactions on Multimedia, 2007, 9(5): 995-1 003.

[12] Po L M, Ma W C. A novel four-step search algorithm for fast block motion estimation[J]. IEEE Transactions on Circuits and Systems for Video Technology, 1996, 6(6): 313-317.

[13] Chow K H K, Liou M L. Generic motion search algorithm for video compression[J]. IEEE Transactions on Circuits and Systems for Video Technology, 1993, 3(12): 440-445.

文章导航

/