H.264并行视频转码算法研究
摘要
视频转码是指对已经压缩编码之后的视频流进行格式上的转换,使经过处理后的视频能满足信道传输或用户终端的要求,其中的格式包括分辨率、码率和编码标准等。由于视频转码需要对已经压缩过的码流进行解码,然后再经过编码转换成满足要求的目标格式码流,因此视频转码是个计算量很大的过程。针对单处理器串行视频转码速度较慢的问题,为了提高视频转码的效率减少转码延迟,本文研究了如何利用并行化的思想来对视频转码进行加速的方法。
本文首先对视频转码技术的系统结构和分类进行了概述,然后介绍了并行算法设计的理论基础并对H.264视频转码的可并行化方法进行了分析,接下来从不同的并行粒度上对H.264的并行视频转码算法进行了研究。论文主要完成以下两个方面的研究工作:
第一,针对现有传统串行转码算法难以解决转码复杂度的问题,本文将基于机群的并行思想应用于H.264视频转码,在构建并行平台的基础上设计了GOP级的H.264并行视频转码算法。该算法以GOP作为并行处理的基本单元,通过构建一个分布式的平台,将分割成段的视频放在平台上来完成并行的视频转码。实验数据表明,该算法能够有效提高视频转码的速度和效率。
第二,为了提高视频转码的效率降低视频转码的算法复杂度,论文根据视频转码的要求和图形处理器的并行结构,提出了一种宏块级的并行算法。该算法利用了GPU强大的并行计算能力,将宏块作为基本处理单元,将复杂的视频转码过程转移到GPU上并行的执行,以达到加速视频转码的目的。实验结果表明,该算法能在保证转码视频质量的条件下将转码速度提高3~4倍,大大缩短了转码延迟,满足了实时性的要求。
Video transcoding is the transformation of converting a compressed video stream to another stream of a different format for adapting to the network or terminal client. These formats include spatial resolution, rate and coding standard. Video transcoding is a complicated process, as it has to decode the input compressed video stream and then encode it into the output format which is in demand. So as to improve the efficiency and reduce the latency of video transcoding, we made this research that how to speed up the transcoding procedure by the theoretic of parallelization.
This paper first introduces the classification and system structure of video transcoding, then introduces the basic theory of parallelization and analyses the possibility of parallelized video transcoding in H.264.In the next the paper researches the method of parallelization of video transcoding in different parallelization levels. The main contributions of this paper are detailed as followed:
First, target to reduce the complexity and the time of video transcoding, we apply the parallelization structure of cluster workstations to the video transcoding of H.264, and design a parallel transcoding method in GOP level. The method takes GOP as the basic processing unit and builds a distributed platform, then transcode the divided videos in the platform synchronously. Experiment results show that this method can greatly improve the transcoding speed and efficiency. Second, so as to improve the efficiency and reduce the latency of video transcoding, we proposed a new method which make full use of the strong parallel computing ability of the graphics processing unit (GPU) based on the parallel structure of GPU. The algorithm takes macroblock as the processing unit and removes the implementation of motion estimation and mode selection which is the most completed part in transcoding procedure from CPU to GPU. Experimental results demonstrate that it can speed up the transcoding process by 3 to 4 times compared as traditional serial algorithm and greatly reduce the transcoding latency.
本文首先对视频转码技术的系统结构和分类进行了概述,然后介绍了并行算法设计的理论基础并对H.264视频转码的可并行化方法进行了分析,接下来从不同的并行粒度上对H.264的并行视频转码算法进行了研究。论文主要完成以下两个方面的研究工作:
第一,针对现有传统串行转码算法难以解决转码复杂度的问题,本文将基于机群的并行思想应用于H.264视频转码,在构建并行平台的基础上设计了GOP级的H.264并行视频转码算法。该算法以GOP作为并行处理的基本单元,通过构建一个分布式的平台,将分割成段的视频放在平台上来完成并行的视频转码。实验数据表明,该算法能够有效提高视频转码的速度和效率。
第二,为了提高视频转码的效率降低视频转码的算法复杂度,论文根据视频转码的要求和图形处理器的并行结构,提出了一种宏块级的并行算法。该算法利用了GPU强大的并行计算能力,将宏块作为基本处理单元,将复杂的视频转码过程转移到GPU上并行的执行,以达到加速视频转码的目的。实验结果表明,该算法能在保证转码视频质量的条件下将转码速度提高3~4倍,大大缩短了转码延迟,满足了实时性的要求。
Video transcoding is the transformation of converting a compressed video stream to another stream of a different format for adapting to the network or terminal client. These formats include spatial resolution, rate and coding standard. Video transcoding is a complicated process, as it has to decode the input compressed video stream and then encode it into the output format which is in demand. So as to improve the efficiency and reduce the latency of video transcoding, we made this research that how to speed up the transcoding procedure by the theoretic of parallelization.
This paper first introduces the classification and system structure of video transcoding, then introduces the basic theory of parallelization and analyses the possibility of parallelized video transcoding in H.264.In the next the paper researches the method of parallelization of video transcoding in different parallelization levels. The main contributions of this paper are detailed as followed:
First, target to reduce the complexity and the time of video transcoding, we apply the parallelization structure of cluster workstations to the video transcoding of H.264, and design a parallel transcoding method in GOP level. The method takes GOP as the basic processing unit and builds a distributed platform, then transcode the divided videos in the platform synchronously. Experiment results show that this method can greatly improve the transcoding speed and efficiency. Second, so as to improve the efficiency and reduce the latency of video transcoding, we proposed a new method which make full use of the strong parallel computing ability of the graphics processing unit (GPU) based on the parallel structure of GPU. The algorithm takes macroblock as the processing unit and removes the implementation of motion estimation and mode selection which is the most completed part in transcoding procedure from CPU to GPU. Experimental results demonstrate that it can speed up the transcoding process by 3 to 4 times compared as traditional serial algorithm and greatly reduce the transcoding latency.
引文
[1] Ahmad Ishfaq, Wei Xiaohui, Sun Yu, Zhang Ya-Qin. Video Transcoding: An Overview of various techniques and Research issue [J]. IEEE Transactions on Multimedia. vol.7, pp.793-804.Oct 2005.
[2] Xin Jun, Lin Chia-Wen, Sun Ming-Ting. Digital Video Transcoding [J]. Proceedings of the IEEE. vol.93, pp.84-97.Jan 2005.
[3]褚晶辉,俞斯乐,鲁照华.视频转换编码及其实现技术的研究[J].电子学报,32(10),2004.
[4] Lonetti F, Martelli F. Motion Vector Composition Algorithm in H.264 Transcoding [J]. 2007 14th Int. Workshop on Systems, Signals and Image Processing, pp.401-404, 2007.
[5] Tan Yap-Peng, Sun Haiwei. Fast motion re-estimation for arbitrary downsizing video transcoding using H.264/Avc standard [J]. IEEE Transactions on Consumer Electronics, 2004,50(3),pp.887-894.
[6] Shi Xun, Yu Xiang, He Dake. Method and device for motion vector estimation in video transcoding using union of search areas. United States Patent Application Publication. 2011.
[7] Liu Yi-Nung, Tang Chi-Sun, Chien Shao-Yi. Coding Mode Analysis of MPEG-2 to H.264/AVC Transcoding for Digital TV Applications [J]. Proceedings - IEEE International Symposium on Circuits and Systems, p 1995-1998, 2007.
[8] Lin Chia-Tien, Teng Wang-Chun, Lin Yinyi. Mode Decision for High-Quality H.264/AVC Video Downscaling Transcoding [J]. Advances in Multimedia Information Processing, pp.252-263, 2010.
[9] Yuh-Reuy Lee, Chia-wen Lin, Cheng-Chien Kao. A DCT-domain Video Transcoder for Spatial Resolution Downconversion [J]. LNCS 2314, pp.207-218, 2002.
[10]杜耀刚,蔡安妮. DCT域视频转码技术综述[J].电子学报,32(10):1644-1648,2005.
[11] J.C, Fernandez, M.p, Malumbres. A Parallel implementation of H. 26L video encoder [J]. Lecture Notes in Computer Science. 2002.
[12]高文,赵德斌,马思伟.数字视频编码技术原理[M].北京:科学出版社,2010.
[13]毕厚杰.新一代视频压缩编码标准—H.264AVC [M].北京:人民邮电出版社,2005.
[14]雷靖.H.264空间分辨率转码研究[D].长沙:湖南大学硕士学位论文,2008.
[15] Shanableh T, Ghanbari M. Heterogeneous Video Transcoding to Lower Spatio-temporal resolutions and different encoding formats [J]. IEEE Trans. Multimedia,2000,(2):101-110.
[16] Bonuccelli, Maurizio, Lonetti, Martelli. Temporal transcoding for mobile video communication [J]. Second Annual International Conference on Mobile and Ubiquitous Systems -Networking and Services, pp.502-506,2005.
[17] Yin P, Wu M, Liu B. Video transcoding by reducing spatial resolution [J]. IEEE International Conference on Image Processing, v 1, p 972-975, 2000.
[18] De Los Reyes, Gustave, Reibman, Amy R, et al. Error-resilient transcoding for video over wireless channels [J]. IEEE Journal on Selected Areas in Communications, v 18, n 6, p 1063-1074, June 2000.
[19] Vetro Antony, Xin Jun, Sun Huifang. Error resilience video trnscoding for wireless communications [J]. International Conference on Wireless Networks and Information Systems, WNIS 2009, p 286-289, 2009.
[20] Sertac Eminsoy, Safak Dogan, Ahmet M.kondoz. Transcoding-based error-resilient video adaptation for3G wireless networks [J]. EURASIP Jounal on Advances in Signal Processing. May,2007.
[21]林杰聪,黄祥林,杨占昕.视频转码技术研究[J].中国传媒大学学报自然科学版,2006,13(3):43-51.
[22]程恺英,王宏远,宋震.视频转码系统集中典型体系结构的比较分析[J].信息终端.2005.
[23] Bjork N, Christopoulos C. Transcoding architectures for Video coding [J]. IEEE Trans. Consumer Electron, 1998, 44(1):88-98.
[24] Shin Yoonjeong, Son, Namrye, Toan Nguyen Dinh, et al. Low-complexity Heterogeneous Video Transcoding [J]. 2010 International Conference on Information Science and Applications, ICISA 2010
[25] Kalva Hari, Kunzelmann Kelly, et al. Reduced Resolution MPEG-2 to H.264 Transcoder [J]. Proceedings of SPIE - The International Society for Optical Engineering, v 7257, 2009.
[26]杨蕾,戴居丰,宋晓炜等.一种MPEG-2到H.264转换编码方案[J].计算及应用,vol.27, 2007.
[27] Rochelle Pereira. Efficient Transcoding of an MPEG-2 bit Stream to an H.264 bit Stream [J].Master of Science in Electrical Engineering, December 2005.
[28] Choi B S, Chun K W, Xin J, et al. An HDTV to SDTV Spatial Transcoder. IEEE Trans. On Circuits and Systems for Video Technology, 2002,12(11):998-1008.
[29] Chih-Hung Li, Chung-Neng Wang, Tihao Chiang. A Fast Downsizing Video Transcoder Based on H.264/AVC Standard [J]. PCM 2004, LNCS 3333, pp.215-233, 2004.
[30] Shen Huifeng, Sun Xiaoyan, Wu feng, et al. A fast downsizing video transcoder for H.264/AVC with rate-distortion optimal mode decision [J]. 2006 IEEE International Conference on Multimedia and Expo, ICME 2006 - Proceedings, v 2006, pp.2017-2020, 2006.
[31]向永生,傅明. H.264视频流分辨率缩减转码的快速宏块模式选择算法[J].计算机应用,vol.28(11), 2008
[32]崔玉斌,蔡安妮. H.264帧间模式选择新算法[J].北京邮电大学报,vol.30(4), 2007.
[33] Zhang P, Lu Y, Huang Q M, Gao W. Mode Mapping Method for H.264 Spatial Downscaling Transcoding [A]. Proc IEEE International Conference on Image Processing. 2004:2781-2784.
[34] Xin Jun, Vetro Antony, Sekiguchi, et al. Motion and mode mapping for MPEG-2 to H.264/AVC transcoding [J]. 2006 IEEE International Conference on Multimedia and Expo, ICME 2006 - Proceedings, v 2006, pp. 313-316, 2006.
[35] Chang S F, Messerschmidt D G. Manipulation and Compositing of MC_DCT Compessed Video. IEEE Journal Selected Area Communications. Vol.13(1):1-11, 1995.
[36] Skodras A N, Christopoulos C. Down-sampling of Compressed Images in the DCT Domain [J]. Proc. European Conference on Signal Processing. 1998. pp.1713-1716.
[37] Shin ll-hong, Lee Yung-L yul, Park Hyunwook. Motion estimation for frame-rate reduction in H.264 transcoding [J]. Proceedings - Second IEEE Workshop on Software Technologies for Future Embedded and Ubiquitous Systems, pp.63-67, 2004.
[38] Hwang Jenq-Neng, Wu Tzong-Der, Lin Chia-wen. Dynamic Fram-skipping in Video Transcoding. IEEE Second Workshop on Multimedia Signal Processing. 1998. pp.616-621.
[39] Youn Jeongnam, Sun Ming-Ting. Fast motion vector composition method for temporal transcoding [J]. Proceedings - IEEE International Symposium on Circuits and Systems, v 4, p IV-243-IV-246, 1999.
[40] Sun H, Kwok W, Zdepski J. W. Architectures for MPEG Compressed Bitstream Scaling [J]. IEEE Transactions on Circuits and Systems for Video Technlolgy, 1996,6(2). Pp.191-199.
[41] Vetro Anthony, Christopoulos Charilaos, Sun H. Video Transcoding Architectures and Techniques: An overview [J]. IEEE Signal Processing Magazine, v 20, n 2, p 18-29, March 2003
[42]张林波,迟学斌,莫则尧等.并行计算导论[M].北京:清华大学出版社.2006.
[43] Kai Hwang, Zhiwei Xu. Scalable Parallel Computing [M],机械工业出版社,1999.
[44] Shameem Akher, Jason Roberts. Multi-Core Programming [M]. Intel Press Corporation. 2006.
[45]蒋兴昌.H.264并行编码算法的研究和优化[D]上海:上海交通大学硕士学位论文,2007..
[46]于俊清,李江,魏海涛.基于同构多核处理器的H.264多粒度并行编码器[J].计算机学报,vol.6, 2009.
[47] Tian Zhiqiang, Xue Jianru, Hu Wei, et al. High Performance Cluster-based Transcoder [J]. ICCASM 2010 International Conference on Computer Application and System Modeling, Proceedings, vol.2, pp.248-252, 2010.
[48]胡亮,蒋春晓,袁本雯.一个支持任务动态调度和进程迁移的PC机群并行计算系统[J].吉林大学学报.No.1, 2002.
[49] Sambe Yasuo, Watanabe,S, Yu D, et al. High-speed distributed video transcoding for multiple rates and formats [J]. IEICE Transactions on Information and Systems, v E88-D, n 8, p 1923-1931, August 2005
[50] Jiani Guo, Fang Chen, Bhuyan, L, Kumar, R. A cluster-based active router architecture supportingvideo-audio stream transcoding service [J]. Parallel and Distributed Processing Symposuim,2003.
[51] Rodriguez A, Gonzales A, Malumbres M P. Performance evaluation of parallel MPEG-4 video coding algorithms on clusters of workstations [J]. International Conference on Parallel Computing in Electrical Engineering: Workshop on System Design Automation, SDA, PARELEC 2004, p 354-357, 2004.
[52]张舒,褚艳丽. GPU高性能运算之CUDA[M].北京:中国水利水电出版社.2009.
[53] Kalva Hari, Colic A, Garcia A, et al. Parallel programming for multimedia applications. Multimedia Tools and Applications, v 51, n 2, p 801-818, January 2011.
[54] Chen Wei-Nien, Hang Hsueh-Ming, H.264/AVC motion estimation implmentation on compute unified device architecture (CUDA) [J]. 2008 IEEE International Conference on Multimedia and Expo, ICME 2008 - Proceedings, p 697-700, 2008.
[55] Pieters Bart, Hollemeerch,C F, et al. Motion estimation for H.264/AVC on multiple GPUs using NVIDIA CUDA [J]. Proceedings of SPIE - The International Society for Optical Engineering, v 7443, 2009.
[56] Jeongnam Youn, Ming-Ting Sum, et al. Motion vector Refinement for High-performance Transcoding [J]. IEEE Transaction on Multimedia vol.1,No.1 Mar 1999.
[2] Xin Jun, Lin Chia-Wen, Sun Ming-Ting. Digital Video Transcoding [J]. Proceedings of the IEEE. vol.93, pp.84-97.Jan 2005.
[3]褚晶辉,俞斯乐,鲁照华.视频转换编码及其实现技术的研究[J].电子学报,32(10),2004.
[4] Lonetti F, Martelli F. Motion Vector Composition Algorithm in H.264 Transcoding [J]. 2007 14th Int. Workshop on Systems, Signals and Image Processing, pp.401-404, 2007.
[5] Tan Yap-Peng, Sun Haiwei. Fast motion re-estimation for arbitrary downsizing video transcoding using H.264/Avc standard [J]. IEEE Transactions on Consumer Electronics, 2004,50(3),pp.887-894.
[6] Shi Xun, Yu Xiang, He Dake. Method and device for motion vector estimation in video transcoding using union of search areas. United States Patent Application Publication. 2011.
[7] Liu Yi-Nung, Tang Chi-Sun, Chien Shao-Yi. Coding Mode Analysis of MPEG-2 to H.264/AVC Transcoding for Digital TV Applications [J]. Proceedings - IEEE International Symposium on Circuits and Systems, p 1995-1998, 2007.
[8] Lin Chia-Tien, Teng Wang-Chun, Lin Yinyi. Mode Decision for High-Quality H.264/AVC Video Downscaling Transcoding [J]. Advances in Multimedia Information Processing, pp.252-263, 2010.
[9] Yuh-Reuy Lee, Chia-wen Lin, Cheng-Chien Kao. A DCT-domain Video Transcoder for Spatial Resolution Downconversion [J]. LNCS 2314, pp.207-218, 2002.
[10]杜耀刚,蔡安妮. DCT域视频转码技术综述[J].电子学报,32(10):1644-1648,2005.
[11] J.C, Fernandez, M.p, Malumbres. A Parallel implementation of H. 26L video encoder [J]. Lecture Notes in Computer Science. 2002.
[12]高文,赵德斌,马思伟.数字视频编码技术原理[M].北京:科学出版社,2010.
[13]毕厚杰.新一代视频压缩编码标准—H.264AVC [M].北京:人民邮电出版社,2005.
[14]雷靖.H.264空间分辨率转码研究[D].长沙:湖南大学硕士学位论文,2008.
[15] Shanableh T, Ghanbari M. Heterogeneous Video Transcoding to Lower Spatio-temporal resolutions and different encoding formats [J]. IEEE Trans. Multimedia,2000,(2):101-110.
[16] Bonuccelli, Maurizio, Lonetti, Martelli. Temporal transcoding for mobile video communication [J]. Second Annual International Conference on Mobile and Ubiquitous Systems -Networking and Services, pp.502-506,2005.
[17] Yin P, Wu M, Liu B. Video transcoding by reducing spatial resolution [J]. IEEE International Conference on Image Processing, v 1, p 972-975, 2000.
[18] De Los Reyes, Gustave, Reibman, Amy R, et al. Error-resilient transcoding for video over wireless channels [J]. IEEE Journal on Selected Areas in Communications, v 18, n 6, p 1063-1074, June 2000.
[19] Vetro Antony, Xin Jun, Sun Huifang. Error resilience video trnscoding for wireless communications [J]. International Conference on Wireless Networks and Information Systems, WNIS 2009, p 286-289, 2009.
[20] Sertac Eminsoy, Safak Dogan, Ahmet M.kondoz. Transcoding-based error-resilient video adaptation for3G wireless networks [J]. EURASIP Jounal on Advances in Signal Processing. May,2007.
[21]林杰聪,黄祥林,杨占昕.视频转码技术研究[J].中国传媒大学学报自然科学版,2006,13(3):43-51.
[22]程恺英,王宏远,宋震.视频转码系统集中典型体系结构的比较分析[J].信息终端.2005.
[23] Bjork N, Christopoulos C. Transcoding architectures for Video coding [J]. IEEE Trans. Consumer Electron, 1998, 44(1):88-98.
[24] Shin Yoonjeong, Son, Namrye, Toan Nguyen Dinh, et al. Low-complexity Heterogeneous Video Transcoding [J]. 2010 International Conference on Information Science and Applications, ICISA 2010
[25] Kalva Hari, Kunzelmann Kelly, et al. Reduced Resolution MPEG-2 to H.264 Transcoder [J]. Proceedings of SPIE - The International Society for Optical Engineering, v 7257, 2009.
[26]杨蕾,戴居丰,宋晓炜等.一种MPEG-2到H.264转换编码方案[J].计算及应用,vol.27, 2007.
[27] Rochelle Pereira. Efficient Transcoding of an MPEG-2 bit Stream to an H.264 bit Stream [J].Master of Science in Electrical Engineering, December 2005.
[28] Choi B S, Chun K W, Xin J, et al. An HDTV to SDTV Spatial Transcoder. IEEE Trans. On Circuits and Systems for Video Technology, 2002,12(11):998-1008.
[29] Chih-Hung Li, Chung-Neng Wang, Tihao Chiang. A Fast Downsizing Video Transcoder Based on H.264/AVC Standard [J]. PCM 2004, LNCS 3333, pp.215-233, 2004.
[30] Shen Huifeng, Sun Xiaoyan, Wu feng, et al. A fast downsizing video transcoder for H.264/AVC with rate-distortion optimal mode decision [J]. 2006 IEEE International Conference on Multimedia and Expo, ICME 2006 - Proceedings, v 2006, pp.2017-2020, 2006.
[31]向永生,傅明. H.264视频流分辨率缩减转码的快速宏块模式选择算法[J].计算机应用,vol.28(11), 2008
[32]崔玉斌,蔡安妮. H.264帧间模式选择新算法[J].北京邮电大学报,vol.30(4), 2007.
[33] Zhang P, Lu Y, Huang Q M, Gao W. Mode Mapping Method for H.264 Spatial Downscaling Transcoding [A]. Proc IEEE International Conference on Image Processing. 2004:2781-2784.
[34] Xin Jun, Vetro Antony, Sekiguchi, et al. Motion and mode mapping for MPEG-2 to H.264/AVC transcoding [J]. 2006 IEEE International Conference on Multimedia and Expo, ICME 2006 - Proceedings, v 2006, pp. 313-316, 2006.
[35] Chang S F, Messerschmidt D G. Manipulation and Compositing of MC_DCT Compessed Video. IEEE Journal Selected Area Communications. Vol.13(1):1-11, 1995.
[36] Skodras A N, Christopoulos C. Down-sampling of Compressed Images in the DCT Domain [J]. Proc. European Conference on Signal Processing. 1998. pp.1713-1716.
[37] Shin ll-hong, Lee Yung-L yul, Park Hyunwook. Motion estimation for frame-rate reduction in H.264 transcoding [J]. Proceedings - Second IEEE Workshop on Software Technologies for Future Embedded and Ubiquitous Systems, pp.63-67, 2004.
[38] Hwang Jenq-Neng, Wu Tzong-Der, Lin Chia-wen. Dynamic Fram-skipping in Video Transcoding. IEEE Second Workshop on Multimedia Signal Processing. 1998. pp.616-621.
[39] Youn Jeongnam, Sun Ming-Ting. Fast motion vector composition method for temporal transcoding [J]. Proceedings - IEEE International Symposium on Circuits and Systems, v 4, p IV-243-IV-246, 1999.
[40] Sun H, Kwok W, Zdepski J. W. Architectures for MPEG Compressed Bitstream Scaling [J]. IEEE Transactions on Circuits and Systems for Video Technlolgy, 1996,6(2). Pp.191-199.
[41] Vetro Anthony, Christopoulos Charilaos, Sun H. Video Transcoding Architectures and Techniques: An overview [J]. IEEE Signal Processing Magazine, v 20, n 2, p 18-29, March 2003
[42]张林波,迟学斌,莫则尧等.并行计算导论[M].北京:清华大学出版社.2006.
[43] Kai Hwang, Zhiwei Xu. Scalable Parallel Computing [M],机械工业出版社,1999.
[44] Shameem Akher, Jason Roberts. Multi-Core Programming [M]. Intel Press Corporation. 2006.
[45]蒋兴昌.H.264并行编码算法的研究和优化[D]上海:上海交通大学硕士学位论文,2007..
[46]于俊清,李江,魏海涛.基于同构多核处理器的H.264多粒度并行编码器[J].计算机学报,vol.6, 2009.
[47] Tian Zhiqiang, Xue Jianru, Hu Wei, et al. High Performance Cluster-based Transcoder [J]. ICCASM 2010 International Conference on Computer Application and System Modeling, Proceedings, vol.2, pp.248-252, 2010.
[48]胡亮,蒋春晓,袁本雯.一个支持任务动态调度和进程迁移的PC机群并行计算系统[J].吉林大学学报.No.1, 2002.
[49] Sambe Yasuo, Watanabe,S, Yu D, et al. High-speed distributed video transcoding for multiple rates and formats [J]. IEICE Transactions on Information and Systems, v E88-D, n 8, p 1923-1931, August 2005
[50] Jiani Guo, Fang Chen, Bhuyan, L, Kumar, R. A cluster-based active router architecture supportingvideo-audio stream transcoding service [J]. Parallel and Distributed Processing Symposuim,2003.
[51] Rodriguez A, Gonzales A, Malumbres M P. Performance evaluation of parallel MPEG-4 video coding algorithms on clusters of workstations [J]. International Conference on Parallel Computing in Electrical Engineering: Workshop on System Design Automation, SDA, PARELEC 2004, p 354-357, 2004.
[52]张舒,褚艳丽. GPU高性能运算之CUDA[M].北京:中国水利水电出版社.2009.
[53] Kalva Hari, Colic A, Garcia A, et al. Parallel programming for multimedia applications. Multimedia Tools and Applications, v 51, n 2, p 801-818, January 2011.
[54] Chen Wei-Nien, Hang Hsueh-Ming, H.264/AVC motion estimation implmentation on compute unified device architecture (CUDA) [J]. 2008 IEEE International Conference on Multimedia and Expo, ICME 2008 - Proceedings, p 697-700, 2008.
[55] Pieters Bart, Hollemeerch,C F, et al. Motion estimation for H.264/AVC on multiple GPUs using NVIDIA CUDA [J]. Proceedings of SPIE - The International Society for Optical Engineering, v 7443, 2009.
[56] Jeongnam Youn, Ming-Ting Sum, et al. Motion vector Refinement for High-performance Transcoding [J]. IEEE Transaction on Multimedia vol.1,No.1 Mar 1999.