基于PC的H.264视频编码器优化设计
摘要
随着信息时代的到来，人们对现有通信网络条件下的多媒体服务提出了更高的要求。例如更高质量的可视电话和视频会议，具有更高清晰度的视频存储和播放，无线网络中的视频通信等。这些应用中一个关键的技术环节是图像的压缩和解压缩。
近年来，图像视频编码技术和编码标准得到了长足的进步。H.264是IEO／IEC和ITU-T两大国际标准化组织联手制定的视频新标准。作为视频编解码领域的最新标准，H.264具有压缩率高、差错恢复能力强、适用范围广等优点。为达到理想的数据压缩率，H.264编码器采用了很多新的编码算法，这些算法极大地增加了编码器的计算复杂性，使其目前在PC环境下的实时应用受到很多限制。要实现H.264的实时编解码，必须对其进行大量的优化工作。但这种优化工作不应该以较大的性能牺牲为代价。
本文以H.264的参考代码JM61e为实验平台，对其编码器进行了详细的分析和性能测试。针对其中导致计算复杂度大大增加的主要因素，采用相应方法改进。在并未显著降低性能的前提下，有效的提高了编码器的效率。
本论文主要包含以下内容：
1、简要介绍了H.264协议的技术亮点和目前的发展状况。尤其对H.264赖以获得高压缩性能的视频编码层中的主要技术做了较详细的介绍。
2、总结了H.264面向网络应用的多种特性和错误恢复方法。
3、详细分析了H.264编码器的性能和算法复杂度，指出影响编码器效率的主要因素。并提出了一种基于率失真最优化准则的快速块匹配算法和基于边缘检测的帧内预测模式选择策略。通过编程实现后，实验显示了改进算法相对于原始算法效率上的明显提升。
4、针对对编码器在PC上的应用做了基于MMX指令集的优化，主要是SAD计算和1／2象素内插部分。
With the coming of IT era, people put forward higher demand on the multimedia service under present condition of communication network. For example, the,higher qulity video phone and video conference; video store and display with higher resolution; video communication in wireless network; etc. But a key technology in all these application is video encode and decode.
Multimedia video coding techonogy and corresponding coding standard are developing rapidly these years. The H.264 is outstanding of them. As the latest standard in video codec field, H.264 has such advantage as high compression rate, strong erroe resilience capacity, highly adaptability and so on. In order to gain good coding result, H.264 adopts so many new algorithms that the amount of calculation increase drastically. As a result, its applications in PC environment are restricted. In order to realize H.264 real-time codec, many optimization works have to be done. But these optimization gains should not at the cost of obviously coding performence drop.
Basded on the reference software--JM61e of H.264 and through a detailed analysis and capability test of the encoder, we find out the major parts that lead to the high computation. After optimized by corresponding method described in this thesis, the efficiency of the encoder is remarkably improved with little performance loss.
The main contents of the thesis are as follow:
1. A brief introduction to the technical highlight of the H.264 standard and its development actuality, especially focus on the main technology in video coding layer which the high compression rate of H.264 is achieved based on.
2. Sammarized the network adaption of H.264 and its error resilience technology which have greatly improved compared to other compression standard such as H.263.
3. Detailedly analyzed the performance and computational complexity of H.264 encoder. Point out the major parts that lead to the high computation. We anvanced a fast block match algorithm based on Rate-Distortion Optimization and a fast intra prediction mode selection strategy based on edge detection. After programming, the result shows greatly improvement of coding efficiency compare to the original algorithm.
4. Find the part that fit to optimization by MMX instruction set and realization them (including the SAD calculation and 1/2 pixel interpolation).
近年来，图像视频编码技术和编码标准得到了长足的进步。H.264是IEO／IEC和ITU-T两大国际标准化组织联手制定的视频新标准。作为视频编解码领域的最新标准，H.264具有压缩率高、差错恢复能力强、适用范围广等优点。为达到理想的数据压缩率，H.264编码器采用了很多新的编码算法，这些算法极大地增加了编码器的计算复杂性，使其目前在PC环境下的实时应用受到很多限制。要实现H.264的实时编解码，必须对其进行大量的优化工作。但这种优化工作不应该以较大的性能牺牲为代价。
本文以H.264的参考代码JM61e为实验平台，对其编码器进行了详细的分析和性能测试。针对其中导致计算复杂度大大增加的主要因素，采用相应方法改进。在并未显著降低性能的前提下，有效的提高了编码器的效率。
本论文主要包含以下内容：
1、简要介绍了H.264协议的技术亮点和目前的发展状况。尤其对H.264赖以获得高压缩性能的视频编码层中的主要技术做了较详细的介绍。
2、总结了H.264面向网络应用的多种特性和错误恢复方法。
3、详细分析了H.264编码器的性能和算法复杂度，指出影响编码器效率的主要因素。并提出了一种基于率失真最优化准则的快速块匹配算法和基于边缘检测的帧内预测模式选择策略。通过编程实现后，实验显示了改进算法相对于原始算法效率上的明显提升。
4、针对对编码器在PC上的应用做了基于MMX指令集的优化，主要是SAD计算和1／2象素内插部分。
With the coming of IT era, people put forward higher demand on the multimedia service under present condition of communication network. For example, the,higher qulity video phone and video conference; video store and display with higher resolution; video communication in wireless network; etc. But a key technology in all these application is video encode and decode.
Multimedia video coding techonogy and corresponding coding standard are developing rapidly these years. The H.264 is outstanding of them. As the latest standard in video codec field, H.264 has such advantage as high compression rate, strong erroe resilience capacity, highly adaptability and so on. In order to gain good coding result, H.264 adopts so many new algorithms that the amount of calculation increase drastically. As a result, its applications in PC environment are restricted. In order to realize H.264 real-time codec, many optimization works have to be done. But these optimization gains should not at the cost of obviously coding performence drop.
Basded on the reference software--JM61e of H.264 and through a detailed analysis and capability test of the encoder, we find out the major parts that lead to the high computation. After optimized by corresponding method described in this thesis, the efficiency of the encoder is remarkably improved with little performance loss.
The main contents of the thesis are as follow:
1. A brief introduction to the technical highlight of the H.264 standard and its development actuality, especially focus on the main technology in video coding layer which the high compression rate of H.264 is achieved based on.
2. Sammarized the network adaption of H.264 and its error resilience technology which have greatly improved compared to other compression standard such as H.263.
3. Detailedly analyzed the performance and computational complexity of H.264 encoder. Point out the major parts that lead to the high computation. We anvanced a fast block match algorithm based on Rate-Distortion Optimization and a fast intra prediction mode selection strategy based on edge detection. After programming, the result shows greatly improvement of coding efficiency compare to the original algorithm.
4. Find the part that fit to optimization by MMX instruction set and realization them (including the SAD calculation and 1/2 pixel interpolation).
引文
[1]ITU-T Recommendation H.261: "Video codec for audiovisual services at p×64 kbit/s," Geneva, 1990, revised at Helsinki, March 1993.
[2]ITU-T and ISO/IEC JTC 1, "Generic coding of moving pictures and associated audio information-Part 2: Video," ITU-T Recommendation H.262-ISO/IEC 13818-2 (MPEG-2),Nov. 1994.
[3]ITU-T Recommendation H.263: "Video coding for low bitrate communication," March 1996.
[4]ITU-T SG16, Gary Sullivan, ed., "Draft text of Recommendation H.263 Version 2("H.263+") for Decision," September 1997.
[5]ISO/IEC/JTC1/SC29/WG11 Doc. N3908, MPEG-4 video verification model ver. 18.0, January 2001
[6]ISO/IEC JTC1, "Coding of audio-visual objects-Part 2: Visual," ISO/IEC 14496-2(MPEG-4 visual version 1), April 1999;Amendment 1(version 2), February, 2000; Amendment 4(streaming profile), January, 2001.
[7]Joint Video Team of ITU-T and ISO/IEC JTC 1, "Draft ITU-T Recommendation and Final Draft International Standard of Joint Video Specification (ITU-T Rec. H.264 I ISO/IEC 14496-10 AVC)," Joint Video Team (JVT) of ISO/IEC MPEG and ITU-T VCEG, JVT-G050, March 2003.
[8]余松煜、张文军、孙军，《现代图像信息压缩技术》，科学出版社，1998，北京
[9]钟玉琢《基于对象的多媒体数据压缩编码国际标准：MPEG-4及其校验模型》科学出版社，2000，北京
[10]《H．264编解码器兴风作浪，意欲动摇MPEG-4的优势地位》http://www. ebnchina.com
[11]何芸 《H．264／AVC标准与市场准备》慧聪广电商务网 http://info. broadcast.sinobnet.com
[12]H．263极低码率窄带电信信道视频编码，朱斌、张春田，电视技术1997年第11期
[13]黄贤武等《数字图象处理与压缩编码技术》电子科技大学出版社
[14]Thomas Wiegand, Gary J. Sullivan, Gisle Bjontegaard, and Ajay Luthra, "Overview of the H.264/AVC Video Coding Standard", IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, JULY 2003
[15]S. Wenger, M. Hannuksela, and T. Stockhammer, "Identified H.26L Applications," ITU-T SG16, Doc. VCEG-L34, Eibsee, Germany, Jan. 2001.
[16]S. Wenger and T.Stockhammer, "H.26L over IP and H.324 Framework," ITU-T VCEG-N52, VCEG (SG16/Q6), Fourteenth Meeting, Santa Barbara, CA, September 2001.
[17]S. Wenger, "H.26L over IP: The IP-Network Adaptation Layer", Proc. Packet Video Workshop 2002, Pittsburgh, PY, April 2002.
[2]ITU-T and ISO/IEC JTC 1, "Generic coding of moving pictures and associated audio information-Part 2: Video," ITU-T Recommendation H.262-ISO/IEC 13818-2 (MPEG-2),Nov. 1994.
[3]ITU-T Recommendation H.263: "Video coding for low bitrate communication," March 1996.
[4]ITU-T SG16, Gary Sullivan, ed., "Draft text of Recommendation H.263 Version 2("H.263+") for Decision," September 1997.
[5]ISO/IEC/JTC1/SC29/WG11 Doc. N3908, MPEG-4 video verification model ver. 18.0, January 2001
[6]ISO/IEC JTC1, "Coding of audio-visual objects-Part 2: Visual," ISO/IEC 14496-2(MPEG-4 visual version 1), April 1999;Amendment 1(version 2), February, 2000; Amendment 4(streaming profile), January, 2001.
[7]Joint Video Team of ITU-T and ISO/IEC JTC 1, "Draft ITU-T Recommendation and Final Draft International Standard of Joint Video Specification (ITU-T Rec. H.264 I ISO/IEC 14496-10 AVC)," Joint Video Team (JVT) of ISO/IEC MPEG and ITU-T VCEG, JVT-G050, March 2003.
[8]余松煜、张文军、孙军，《现代图像信息压缩技术》，科学出版社，1998，北京
[9]钟玉琢《基于对象的多媒体数据压缩编码国际标准：MPEG-4及其校验模型》科学出版社，2000，北京
[10]《H．264编解码器兴风作浪，意欲动摇MPEG-4的优势地位》http://www. ebnchina.com
[11]何芸 《H．264／AVC标准与市场准备》慧聪广电商务网 http://info. broadcast.sinobnet.com
[12]H．263极低码率窄带电信信道视频编码，朱斌、张春田，电视技术1997年第11期
[13]黄贤武等《数字图象处理与压缩编码技术》电子科技大学出版社
[14]Thomas Wiegand, Gary J. Sullivan, Gisle Bjontegaard, and Ajay Luthra, "Overview of the H.264/AVC Video Coding Standard", IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, JULY 2003
[15]S. Wenger, M. Hannuksela, and T. Stockhammer, "Identified H.26L Applications," ITU-T SG16, Doc. VCEG-L34, Eibsee, Germany, Jan. 2001.
[16]S. Wenger and T.Stockhammer, "H.26L over IP and H.324 Framework," ITU-T VCEG-N52, VCEG (SG16/Q6), Fourteenth Meeting, Santa Barbara, CA, September 2001.
[17]S. Wenger, "H.26L over IP: The IP-Network Adaptation Layer", Proc. Packet Video Workshop 2002, Pittsburgh, PY, April 2002.