• DocumentCode
    2645697
  • Title

    A Novel H.264-based Multiple Description Video Coding Via Polyphase Transform and Partial Prediction

  • Author

    Wei, Zhe ; Cai, Canhui ; Ma, Kai-Kuang

  • Author_Institution
    Inst. of Inf. Sci. & Eng., Huaqiao Univ., Fujian
  • fYear
    2006
  • fDate
    12-15 Dec. 2006
  • Firstpage
    151
  • Lastpage
    154
  • Abstract
    In this paper, a new H.264-based multiple description coding (MDC) scheme is proposed and called the prediction-based spatial polyphase transform (PSPT) MDC, to enhance the error resilience and reduce the computational complexity of the H.264 codec. In our approach, each input frame is first subsampled in both horizontal and vertical directions to form four subframes. Instead of directly coding and transporting all these subframes, two of them are predicted based on their neighboring subframes, followed by encoding and transmission. To handle the prediction process between the adjacent subframes, a hybrid interpolation algorithm is proposed, which combines the bilinear interpolation with some gradient tests. The simulation results demonstrate that the proposed MDC scheme is more robust and yields better reconstructed video qualities than existing approaches under an error-prone condition. Moreover, the computational complexity of the proposed encoder is reduced by half approximately, compared with that of the H.264/AVC
  • Keywords
    computational complexity; interpolation; video coding; H.264 codec; bilinear interpolation; computational complexity reduction; error resilience; hybrid interpolation algorithm; multiple description video coding; partial prediction; polyphase transform; subframes; video qualities; Automatic voltage control; Codecs; Computational complexity; Computational modeling; Encoding; Interpolation; Resilience; Robustness; Testing; Video coding;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Intelligent Signal Processing and Communications, 2006. ISPACS '06. International Symposium on
  • Conference_Location
    Yonago
  • Print_ISBN
    0-7803-9732-0
  • Electronic_ISBN
    0-7803-9733-9
  • Type

    conf

  • DOI
    10.1109/ISPACS.2006.364856
  • Filename
    4212243