• DocumentCode
    3316020
  • Title

    Adaptive motion vector resolution with implicit signaling

  • Author

    Guo, Liwei ; Yin, Peng ; Zheng, Yunfei ; Lu, Xiaoan ; Xu, Qian ; Sole, Joel

  • Author_Institution
    Technicolor Corp. Res., Princeton, NJ, USA
  • fYear
    2010
  • fDate
    26-29 Sept. 2010
  • Firstpage
    2057
  • Lastpage
    2060
  • Abstract
    In all video coding standards, motion vectors are restricted to have the same resolution for the whole sequence. In this paper, we propose an adaptive motion vector resolution scheme to improve the coding efficiency. The scheme is based on an analytical model of motion compensated residue which studies the relationship between motion vector errors and the variance of the residue. With this model, the proposed scheme can adaptively select motion vector resolution with implicit signaling. Experimental results show that the proposed scheme can improve coding efficiency at slightly increased encoding complexity.
  • Keywords
    video coding; adaptive motion vector resolution; analytical model; implicit signaling; motion vectors; video coding standards; Adaptation model; Decoding; Encoding; Image coding; Image resolution; Motion estimation; Signal resolution; adaptive motion vector resolution; motion compensation; signal modeling; video coding;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Image Processing (ICIP), 2010 17th IEEE International Conference on
  • Conference_Location
    Hong Kong
  • ISSN
    1522-4880
  • Print_ISBN
    978-1-4244-7992-4
  • Electronic_ISBN
    1522-4880
  • Type

    conf

  • DOI
    10.1109/ICIP.2010.5650435
  • Filename
    5650435