• DocumentCode
    3498665
  • Title

    A Rate Control Scheme Based on MAD Weighted Model for H.264/AVC

  • Author

    Ruan Ruolin ; Hu Ruimin ; Xiao Xuqing

  • Author_Institution
    Nat. Eng. Res. Center for Multimedia Software, Wuhan Univ., Wuhan
  • fYear
    2007
  • fDate
    21-25 Sept. 2007
  • Firstpage
    2903
  • Lastpage
    2906
  • Abstract
    Under the network bandwidth and the delay constrained, the rate control has become a key technique for video coding in order to obtain consecutive and high quality reconstructive video picture. Rate control scheme of the basic unit in H.264/AVC mainly adopted the linear MAD predict model and the quadratic rate distortion model, in the process of implementing, after coding a macroblock, the parameters of the models will be updated, and then computes the quantization parameters of the current macroblock. So, its computation cost is very high, and its complexity is also very high. Through analyzing the rate control scheme of the H.264/AVC JVT G012rl, the paper proposed an improved low complexity the MAD weighted predict model, and make the accurate rate control in the macroblock layer, and carried out it in the JM98 platform of JVT reference software in H.264/AVC. Extensive experiment results show the complexity of this scheme is lower than the JVT-G012rl of H.264/AVC, and the average PSNR of the usually standard test sequences increased O.lldB, at the same time, its accuracy of the rate control of the QCIF sequences averagely improved 0.498 kpbs.
  • Keywords
    video coding; H.264/AVC; mean absolute difference; network bandwidth; quadratic rate distortion model; rate control scheme; video coding; Automatic voltage control; Bandwidth; Bit rate; Computational efficiency; PSNR; Predictive models; Quantization; Rate-distortion; Software quality; Video coding;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Wireless Communications, Networking and Mobile Computing, 2007. WiCom 2007. International Conference on
  • Conference_Location
    Shanghai
  • Print_ISBN
    978-1-4244-1311-9
  • Type

    conf

  • DOI
    10.1109/WICOM.2007.721
  • Filename
    4340496