A Fast Scheme for Converting DCT Coefficients to H.264/AVC Integer Transform Coefficients

  • Gao Chen
  • Shouxun Lin
  • Yongdong Zhang
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4141)


Converting MPEG-2 8-tap discrete cosine transform (DCT) coefficients to H.264/AVC 4-tap integer transform (IT) coefficients is one of the indispensable operations for MPEG-2 to H.264 transform domain transcoding. In this paper, we propose a novel transform kernel matrix based on the factorization of DCT to perform this operation directly in transform domain. Furthermore, additional reduction in computation can be obtained by taking advantage of the fact that most of the 8×8 DCT blocks in real video sequences only have nonzero coefficients in the upper left 4×4 quadrant. Relative to Jun Xin’s method, the proposed method saves 16 operations in the worst case and 440 operations in the best case for each 8×8 DCT block. Experimental results also show that the proposed method efficiently reduces the computational cost up to 42.9% with negligible degradation in video quality. Hence, the proposed method can be efficiently used in the real-time and high capacity MPEG-2 to H.264 transform domain transcoding.


Discrete Cosine Transform Coefficient Conversion Discrete Cosine Transform Coefficient Inverse Discrete Cosine Transform Pixel Domain 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Wiegand, T., Sullivan, G.: Draft ITU-T recommendation and final draft international standard of joint video specification (ITUT Rec. H.264 | ISO/IEC 14496-10 AVC), JVT-G050, Pattaya, Thailand (March 2003)Google Scholar
  2. 2.
    ISO/IEC JTC11/SC29/WG11, Generic coding of moving pictures and associated audio iformation: vdeo, ISO/IEC 13818-2 (May 1994), Netravali, A.N., Haskell, B.G.: Digital Pictures, 2nd edn., pp. 613–651. Plenum Press, New York (1995)Google Scholar
  3. 3.
    Vetro, A., Christopoulos, C., Sun, H.: Video tanscoding achitectures and tehniques: an overiew. IEEE Signal Processing Magazine 20(2), 18–29 (2003)CrossRefGoogle Scholar
  4. 4.
    Kalva, H.: Issues in H.264/MPEG-2 video transcoding. In: CCNC 2004, First IEEE, January 5-8 (2004)Google Scholar
  5. 5.
    Xin, J., Vetro, A., Sun, H.: Converting DCT coefficients to H.264/AVC transform coefficients. In: Aizawa, K., Nakamura, Y., Satoh, S. (eds.) PCM 2004. LNCS, vol. 3332, pp. 939–946. Springer, Heidelberg (2004)CrossRefGoogle Scholar
  6. 6.
    Shen, B.: From 8-tap DCT to 4-tap integer-transform for MPEG to H.264 transcoding. In: Proc. IEEE International Conf. on Image Processing (ICIP 2004) (October 2004)Google Scholar
  7. 7.
    Arai, Y., Agui, T., Nakajima, M.: A fast DCT-SQ scheme for images. Trans. Of the IEICE E 71(11), 1095 (1988)Google Scholar
  8. 8.
    Pennebaker, W.B., Mitchell, J.L.: JPEG still image data compression standard, pp. 53–57. Van Nostrand Reinhold (1993) Google Scholar
  9. 9.
    MPEG-2 video decodec v12, available online at:
  10. 10.
    H.264/AVC reference software JM8.2, available online at:

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Gao Chen
    • 1
  • Shouxun Lin
    • 1
  • Yongdong Zhang
    • 1
  1. 1.Institute of Computing TechnologyChinese Academy of SciencesBeijingP.R. China

Personalised recommendations