• DocumentCode
    2098232
  • Title

    Search of Fast Mode Selection Algorithm Based on Macroblock-Adaptive

  • Author

    Wei-xing, Zhu ; Ri-xia, Wang ; Xin-cheng, Li

  • Author_Institution
    Sch. of Electr. & Inf. Eng., Jiangsu Univ., Zhenjiang, China
  • Volume
    2
  • fYear
    2008
  • fDate
    20-22 Dec. 2008
  • Firstpage
    485
  • Lastpage
    488
  • Abstract
    Based on the statistical analysis of relation between CBP (Coded Block Pattern), MVD (Motion Vector Difference) and each coding mode, a fast mode selection algorithm based on macroblock-adaptive is proposed. In this algorithm, adaptive judgment strategies are adopted to filter partial coding mode quickly. The residual CBP, MVD information is used adequately to identify the relativity of intra and inter frame. It only needs to compute the radio distortion cost of mode with very high possibility according to judgment rule combining with RD cost. Experimental results show that at the expense of 3.8 percent increase of bit rate, the coding complexity is reduced and the efficiency of coding mode selection is increased greatly. The encoding speed and the PSNR (peak signal to noise ratio) increases 53.27% and 0.023 dB respectively in average.
  • Keywords
    block codes; code standards; computational complexity; correlation methods; data compression; distortion; filtering theory; image sequences; motion estimation; optimisation; statistical analysis; video coding; CBP information coding; H.264 code; coded block pattern; coding complexity; correlation method; fast mode selection algorithm; filter partial coding mode; intra mode selection; macroblock-adaptive; motion estimation; motion vector difference; radio distortion optimization; statistical analysis; video sequence; Bit rate; Corporate acquisitions; Cost function; Image coding; Image quality; Information filtering; Information filters; Motion analysis; PSNR; Statistical analysis; H.264/AVC; coded block pattern; fast mode decision; macroblock mode; motion vector;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computer Science and Computational Technology, 2008. ISCSCT '08. International Symposium on
  • Conference_Location
    Shanghai
  • Print_ISBN
    978-1-4244-3746-7
  • Type

    conf

  • DOI
    10.1109/ISCSCT.2008.279
  • Filename
    4731669