MPEG-2到H.264的转码中关键技术的研究
摘要
通信、计算机和互联网的结合为视频应用带来了前所未有的机遇但也对压缩视频的传输提出了新的挑战。多种视频压缩标准的存在,导致视频设备间的兼容问题,而视频转码是有效解决这种兼容问题的一种技术。H.264作为新一代的视频压缩标准,它具有在相同图像质量下比其它视频压缩标准拥有更高的压缩率的特性,而当前保留的大多视频节目都是由MPEG-2压缩得到的,所以为了在相同的空间存储更多的视频节目,由MPEG-2到H.264的视频转换是很有必要的。本文研究工作就是围绕视频转码技术展开的。
本文的研究目标是在保证图像质量相对于MPEG-2压缩的码流没有大的降低的情况下降低更多的码率。首先介绍了转码技术的发展和研究方向以及视频压缩的基本原理,然后分析了H.264视频压缩标准的基本结构和流程,并介绍H.264协议的先进技术,通过介绍这些特点分析转码的必要性。在细致的分析了MPEG-2解码器和H.264编码器后,使用了重用MPEG-2中的MV和QP值的转码算法,通过此算法有助于提高转码后图像的质量并且码率的增加较少。在MV重用的算法中运用了运动矢量的修正和重构的方法;在QP值的重用的算法中使用了一种增大B帧的QP值已达到在更低的码率有更好的图像质量的算法。在基本相同的码率的情况下比较了本文的转码方案和直接级联模式的性能,得出本文方案比直接级联模式PSNR高0.1～0.4dB的结论。然后再将本文转码方案的效果同在MPEG-2下压缩的视频码流作比较,得出在降低图像质量0.2～0.5dB的情况下可节约20%～30%的码率的转码结果。还介绍了一种可用于本文转码的H.264快速模式抉择优化方案,此优化方案在相对原始的编码器节约40%左右的时间的情况下,图像质量只有0.1dB左右的下降,码率只有2%～4%的增大的优化效果。此方案同样可以运用在H.264的编码器上。
The development of multimedia systems brings a major influence in the area of image and video encoding.The interactivity and integration of video data with computer, cellular and television systems becomes a new problem of research worldwide.As the content representation formats increase progressively,interoperability between different systems is becoming more and more important. Video content transcoding is one of the key technology to solve the problem. H.264 as the new video compressing standard, has a better compressing ability than former standards both in video quality and compression rate. Since the most video programs are currently compressed by the standard of MPEG-2, the transcoding between MPEG-2 and H.264 is a vital issue. This thesis is focus on the research of video transcoding.
This thesis is goal to reduce bitrate without decreasing the video quality obviously compared with the orignal video compressed by MPEG-2.The structure of this thesis is as following:
Firstly the thesis introduces the development of transcoding and the basic theory of video compressing. Then analyses the structure and the flow of the H.264 encoder and MPEG-2 decoder. After that, this thesis uses an algorithm of MV and QP reuse. With the used algorithm, we can get 0.1~0.4dB gain in the PSNR at the same bitrate, compared with the CPDT algorithm, which just connects the MPEG-2 decoder and H.264 encoder directly. This proposal method can save bitrate by 20%~30% with only 0.2~0.5dB decreasing in PSNR,compared with the MPEG-2 encoded stream(The PSNRs of MPEG-2 encoded stream and transcoded stream are both calculated based on the same original yuv sequence). Finaly, the thesis introduces a method to optimize the mode decision of the transcoder. It can save transcoding time by 40% or so, with only 0.1dB decrease in PSNR and 2%~4% increase in bitrate. This method can also be uesd in H.264 encoder.
本文的研究目标是在保证图像质量相对于MPEG-2压缩的码流没有大的降低的情况下降低更多的码率。首先介绍了转码技术的发展和研究方向以及视频压缩的基本原理,然后分析了H.264视频压缩标准的基本结构和流程,并介绍H.264协议的先进技术,通过介绍这些特点分析转码的必要性。在细致的分析了MPEG-2解码器和H.264编码器后,使用了重用MPEG-2中的MV和QP值的转码算法,通过此算法有助于提高转码后图像的质量并且码率的增加较少。在MV重用的算法中运用了运动矢量的修正和重构的方法;在QP值的重用的算法中使用了一种增大B帧的QP值已达到在更低的码率有更好的图像质量的算法。在基本相同的码率的情况下比较了本文的转码方案和直接级联模式的性能,得出本文方案比直接级联模式PSNR高0.1～0.4dB的结论。然后再将本文转码方案的效果同在MPEG-2下压缩的视频码流作比较,得出在降低图像质量0.2～0.5dB的情况下可节约20%～30%的码率的转码结果。还介绍了一种可用于本文转码的H.264快速模式抉择优化方案,此优化方案在相对原始的编码器节约40%左右的时间的情况下,图像质量只有0.1dB左右的下降,码率只有2%～4%的增大的优化效果。此方案同样可以运用在H.264的编码器上。
The development of multimedia systems brings a major influence in the area of image and video encoding.The interactivity and integration of video data with computer, cellular and television systems becomes a new problem of research worldwide.As the content representation formats increase progressively,interoperability between different systems is becoming more and more important. Video content transcoding is one of the key technology to solve the problem. H.264 as the new video compressing standard, has a better compressing ability than former standards both in video quality and compression rate. Since the most video programs are currently compressed by the standard of MPEG-2, the transcoding between MPEG-2 and H.264 is a vital issue. This thesis is focus on the research of video transcoding.
This thesis is goal to reduce bitrate without decreasing the video quality obviously compared with the orignal video compressed by MPEG-2.The structure of this thesis is as following:
Firstly the thesis introduces the development of transcoding and the basic theory of video compressing. Then analyses the structure and the flow of the H.264 encoder and MPEG-2 decoder. After that, this thesis uses an algorithm of MV and QP reuse. With the used algorithm, we can get 0.1~0.4dB gain in the PSNR at the same bitrate, compared with the CPDT algorithm, which just connects the MPEG-2 decoder and H.264 encoder directly. This proposal method can save bitrate by 20%~30% with only 0.2~0.5dB decreasing in PSNR,compared with the MPEG-2 encoded stream(The PSNRs of MPEG-2 encoded stream and transcoded stream are both calculated based on the same original yuv sequence). Finaly, the thesis introduces a method to optimize the mode decision of the transcoder. It can save transcoding time by 40% or so, with only 0.1dB decrease in PSNR and 2%~4% increase in bitrate. This method can also be uesd in H.264 encoder.
引文
[1]钟玉琢,王琪,赵黎,杨小勤 编译.MPEG-2 运动图像压缩编码国际标准及 MPEG 的新进展 北京:清华大学出版社,2002,158-162,206-207
[2]毕厚杰.新一代视频压缩编码标准-H.264/AVC. 北京:人民邮电出版社,2005,22-24,132-135
[3]干宗良,李小蕾.H.264 的变换编码和量化过程分析.电视技术,2003,258(12):7-9
[4]程恺英,王宏远,樊淳标.数字视频转码技术综述.数字电视与数字视频,2005,274(4):13-16
[5]肖友能,薛向阳,曾玮.视频转码技术回顾.通信学报,2002,23(8):72-80
[6]石磊,楼剑,虞露.基于应用的转码过程中的QP映射. 中国图象图形学报,2005,10(6):779-784
[7] Zhi Zhou,Shijun SunShawmin Lei,Ming-Ting Sun.Motion Information and Coding Mode Reuse for MPEG-2 to H.264 Transcoding. Circuit and systems.2005.ISCAS2005.IEEE international Symposium on 23-26 May 2005 vol.2 Page(s):1230-1233
[8] Tuukka Toivonen,Janne Heikkil.Reduced Frame Quantization in Video Coding. 2006 Springer-Verlag http://www.springeronline.com/lncs.Published in VLBV 2005
[9]Joint Video Team of ITU-T and ISO/IEC JTC 1, ”ITU-T Recommendation and International Standard of Joint Video Specification (ITU-T Rec. H.264 | ISO/IEC 14496-10 AVC),”March 2005.
[10]Joint Video Team (JVT) of ISO/IEC MPEG & ITU-T VCEG(ISO/IEC JTC1/SC29/WG11 and ITU-T SG16 Q.6)6th Meeting: Awaji, Island, JP, 5-13 December,2002
[11]Joint Video Team (JVT) of ISO/IEC MPEG and ITU-T VCEG Thailand, Jan 2006
[12]FEAMSTER N,W EE S J An MPEG·2 to H.263 transcoder[A].SPIE Voice,Video,and Data Communications Conference[C].Boston,September 1999.352~360
[13]朱冬冬,戴琼海.H.264快速帧间编码模式选择算法.有线电视,2004,153(9):36-40
[14]王辅中,丁嵘,戴琼海.一种基于视频转码的IP电视应用.有线电视技术,2005,172(4):48-50
[15]李植荣,余松煜.减少码率变换再量化误差的优化方案. 红外与激光工程,2002,31(5):455-459
[16] Xiaoan Lu, Alexis Michael Tourapis, Peng Yin ,Jill Boyce.Fast Mode Decision and MotionEstimation for H.264 with a Focus on MPEG-2/H.264 Transcoding. Circuit and systems.2005.IS CAS2005. IEEE international Symposium on. 23-26 May 2005 . Vol. 2 Page(s):1246 - 1249
[17] P. Yin, H.-Y.C. Tourapis, A.M. Tourapis, and J. Boyce, “Fast modedecision and motion estimation for JVT/H.264,” in Proceedings of International Conference on Image Processing, 2003, pp. 853-856
[18]Zhi Zhou, Ming-Ting Sun ,Yuh-Feng Hsu, “Fast variable blocksizemotion estimation based on merging and splitting procedures for H.264/MPEG-4 AVC,” IEEE Intl. Symposium on Circuits andSystems (ISCAS) 2004, Vancouver, BC, Canada, May 2004,pp.305~309
[19] Susie J. Wee, John G. Apostolopoulos, and Nick Feamster, “Field-toframe transcoding with spatial and temporal downsampling,” IEEE Intl. Conf. Image Processing, 1999, Vol. 4, pp. 271-275
[20]Aturl Puri,Xuemin Chen,Ajay Luthra.Video coding using the H.264/MPEG-4 AVC compression standard . Signal Processing: Image Communication, v 19, n 9 SPEC. ISS., October, 2004,p:793-849
[21]王辅中,戴琼海,丁 嵘.视频转码中的运动重估计技术.有线电视技术,2004,162(18):24-28
[22]N.Feamster,C.Christopoulos.transcoder architectures for video coding.IEEE Trans.Consumer Electron,vol.44,Feb.1998,pp 88～98
[23] Luthra,Ajay ,Topiwala.Pankai. Overview of the H.264/AVC video coding standard. Proceedings of SPIE - The International Society for Optical Engineering, v 5203, Applications of Digital Image Processing XXVI, 2003, p 417-431
[24]P.Assuncao and M.Ghanbari,A frequency-domain video transcoder for dynamic bit-rate redu ction of MPEG-2 bitstreams, IEEE Trans. Circuit Syst Video Technol,vol 8,pp 953-967,Dec.1998
[25]N.Bjork and C.Christopoulos,Transcoder architectures for video coding ,IEEE Trans .consumer Electron ,vol.44,Feb 1998,pp 88-98
[26]J.R.Yee and B.L.Yeo ,A fast algorithm for DCT-domain inverse motion compensation based on shared information in a macroblock, IEEE Trans Circuits Syst Video Technol,vol.10 pp 767-775 Aug.2000
[27]S.-H Jang and N.S.Jayant,An adaptive nonlinear motion vector resampling algorithm for down -scaling video transcoding,in proc.IEEE conf.multimedia and Expo,vol.2 ,2003,pp. 229-232
[28]N.Werner ,Requantization for transcoding of MPEG-2 intraframes,IEEE Trans .image Process, vol.8 no.2,pp.179-191,Feb.1999
[29]A.vetro,H.sun,and Y.Wang, Objec-based transcoding for adaptive video content delivery,IEEE Trans .Circuit Syst.Video technol,vol.11,pp.387-401,Mar.2001
[30]A.vetro,T.Hata,N.Kuwahara,H.kalva,and S.Sekiguchi,Complexity-quality evalution of trans_ coding architectures for reduced spatial resolution,IEEE Trans.Consumer Electron,vol.48,pp. 515-521,Aug.2002
[31]潘建,数字视频编辑及网络传输中若干问题研究,上海:上海交通大学博士学位论文,1999年6月
[32]Nuno Roma,Leonel Sousa,Fast Transcoding Architectures for insertion of non-regular shaped objects in the compressed DCT domain,Signal Processing:Image Commuication,Sep.2003,vol.18(8), pp.659-683
[33]P.Assuncao and M.Ghanbari,Buffer analysis and control in CBR video trancoding ,IEEE Trans. Circuits Syst.Video Technol,vol.10,pp.83-92,Feb.2000
[34]P.Yin,A.Vetro,B.Lui,and H.Sun,Drift compensation for reduced spatial resolution transcoding , IEEE Trans .circuits Syst .Video Technol ,vol.12,pp.1009-1020,Nov.2002
[35]J.Xin,M.-T.Sun,B.S.Choi,and K.W.Chun,An HDTV to SDTV spatial transcoder,IEEE Trans. Circuit Syst.Video Technol,vol.12,no.11,pp.998-1008,Nov.2002
[2]毕厚杰.新一代视频压缩编码标准-H.264/AVC. 北京:人民邮电出版社,2005,22-24,132-135
[3]干宗良,李小蕾.H.264 的变换编码和量化过程分析.电视技术,2003,258(12):7-9
[4]程恺英,王宏远,樊淳标.数字视频转码技术综述.数字电视与数字视频,2005,274(4):13-16
[5]肖友能,薛向阳,曾玮.视频转码技术回顾.通信学报,2002,23(8):72-80
[6]石磊,楼剑,虞露.基于应用的转码过程中的QP映射. 中国图象图形学报,2005,10(6):779-784
[7] Zhi Zhou,Shijun SunShawmin Lei,Ming-Ting Sun.Motion Information and Coding Mode Reuse for MPEG-2 to H.264 Transcoding. Circuit and systems.2005.ISCAS2005.IEEE international Symposium on 23-26 May 2005 vol.2 Page(s):1230-1233
[8] Tuukka Toivonen,Janne Heikkil.Reduced Frame Quantization in Video Coding. 2006 Springer-Verlag http://www.springeronline.com/lncs.Published in VLBV 2005
[9]Joint Video Team of ITU-T and ISO/IEC JTC 1, ”ITU-T Recommendation and International Standard of Joint Video Specification (ITU-T Rec. H.264 | ISO/IEC 14496-10 AVC),”March 2005.
[10]Joint Video Team (JVT) of ISO/IEC MPEG & ITU-T VCEG(ISO/IEC JTC1/SC29/WG11 and ITU-T SG16 Q.6)6th Meeting: Awaji, Island, JP, 5-13 December,2002
[11]Joint Video Team (JVT) of ISO/IEC MPEG and ITU-T VCEG Thailand, Jan 2006
[12]FEAMSTER N,W EE S J An MPEG·2 to H.263 transcoder[A].SPIE Voice,Video,and Data Communications Conference[C].Boston,September 1999.352~360
[13]朱冬冬,戴琼海.H.264快速帧间编码模式选择算法.有线电视,2004,153(9):36-40
[14]王辅中,丁嵘,戴琼海.一种基于视频转码的IP电视应用.有线电视技术,2005,172(4):48-50
[15]李植荣,余松煜.减少码率变换再量化误差的优化方案. 红外与激光工程,2002,31(5):455-459
[16] Xiaoan Lu, Alexis Michael Tourapis, Peng Yin ,Jill Boyce.Fast Mode Decision and MotionEstimation for H.264 with a Focus on MPEG-2/H.264 Transcoding. Circuit and systems.2005.IS CAS2005. IEEE international Symposium on. 23-26 May 2005 . Vol. 2 Page(s):1246 - 1249
[17] P. Yin, H.-Y.C. Tourapis, A.M. Tourapis, and J. Boyce, “Fast modedecision and motion estimation for JVT/H.264,” in Proceedings of International Conference on Image Processing, 2003, pp. 853-856
[18]Zhi Zhou, Ming-Ting Sun ,Yuh-Feng Hsu, “Fast variable blocksizemotion estimation based on merging and splitting procedures for H.264/MPEG-4 AVC,” IEEE Intl. Symposium on Circuits andSystems (ISCAS) 2004, Vancouver, BC, Canada, May 2004,pp.305~309
[19] Susie J. Wee, John G. Apostolopoulos, and Nick Feamster, “Field-toframe transcoding with spatial and temporal downsampling,” IEEE Intl. Conf. Image Processing, 1999, Vol. 4, pp. 271-275
[20]Aturl Puri,Xuemin Chen,Ajay Luthra.Video coding using the H.264/MPEG-4 AVC compression standard . Signal Processing: Image Communication, v 19, n 9 SPEC. ISS., October, 2004,p:793-849
[21]王辅中,戴琼海,丁 嵘.视频转码中的运动重估计技术.有线电视技术,2004,162(18):24-28
[22]N.Feamster,C.Christopoulos.transcoder architectures for video coding.IEEE Trans.Consumer Electron,vol.44,Feb.1998,pp 88～98
[23] Luthra,Ajay ,Topiwala.Pankai. Overview of the H.264/AVC video coding standard. Proceedings of SPIE - The International Society for Optical Engineering, v 5203, Applications of Digital Image Processing XXVI, 2003, p 417-431
[24]P.Assuncao and M.Ghanbari,A frequency-domain video transcoder for dynamic bit-rate redu ction of MPEG-2 bitstreams, IEEE Trans. Circuit Syst Video Technol,vol 8,pp 953-967,Dec.1998
[25]N.Bjork and C.Christopoulos,Transcoder architectures for video coding ,IEEE Trans .consumer Electron ,vol.44,Feb 1998,pp 88-98
[26]J.R.Yee and B.L.Yeo ,A fast algorithm for DCT-domain inverse motion compensation based on shared information in a macroblock, IEEE Trans Circuits Syst Video Technol,vol.10 pp 767-775 Aug.2000
[27]S.-H Jang and N.S.Jayant,An adaptive nonlinear motion vector resampling algorithm for down -scaling video transcoding,in proc.IEEE conf.multimedia and Expo,vol.2 ,2003,pp. 229-232
[28]N.Werner ,Requantization for transcoding of MPEG-2 intraframes,IEEE Trans .image Process, vol.8 no.2,pp.179-191,Feb.1999
[29]A.vetro,H.sun,and Y.Wang, Objec-based transcoding for adaptive video content delivery,IEEE Trans .Circuit Syst.Video technol,vol.11,pp.387-401,Mar.2001
[30]A.vetro,T.Hata,N.Kuwahara,H.kalva,and S.Sekiguchi,Complexity-quality evalution of trans_ coding architectures for reduced spatial resolution,IEEE Trans.Consumer Electron,vol.48,pp. 515-521,Aug.2002
[31]潘建,数字视频编辑及网络传输中若干问题研究,上海:上海交通大学博士学位论文,1999年6月
[32]Nuno Roma,Leonel Sousa,Fast Transcoding Architectures for insertion of non-regular shaped objects in the compressed DCT domain,Signal Processing:Image Commuication,Sep.2003,vol.18(8), pp.659-683
[33]P.Assuncao and M.Ghanbari,Buffer analysis and control in CBR video trancoding ,IEEE Trans. Circuits Syst.Video Technol,vol.10,pp.83-92,Feb.2000
[34]P.Yin,A.Vetro,B.Lui,and H.Sun,Drift compensation for reduced spatial resolution transcoding , IEEE Trans .circuits Syst .Video Technol ,vol.12,pp.1009-1020,Nov.2002
[35]J.Xin,M.-T.Sun,B.S.Choi,and K.W.Chun,An HDTV to SDTV spatial transcoder,IEEE Trans. Circuit Syst.Video Technol,vol.12,no.11,pp.998-1008,Nov.2002