• DocumentCode
    2975486
  • Title

    Statistical data based low-complexity mode decision for P-frame encoding in H.264/AVC

  • Author

    Bin Zhan ; Hou, Baochun ; Sotudeh, Reza

  • Author_Institution
    Univ. of Hertfordshire, Hatfield
  • fYear
    2007
  • fDate
    10-13 Dec. 2007
  • Firstpage
    1
  • Lastpage
    5
  • Abstract
    Compared with former video coding standards, H.264/AVC introduces a number of impressive techniques to improve the video coding efficiency. Flexible block size based mode decision (MD) is one of the major improvements and enables the encoder to accurately and efficiently predict the detailed contents in each frame. However, extremely complex calculations are involved in MD, and significantly increased consumption of power and time is also incurred during the encoding processing. This paper aims at presenting a simple but efficient fast MD algorithm. The proposed algorithm reduces the number of candidate inter modes according to the statistical data obtained from the previously encoded frames. A bitrate based adjustment scheme is also introduced in order to avoid the degradation of encoding efficiency. An improvement of encoding time with slightly decreased performance has been achieved, the experimental results are also provided.
  • Keywords
    statistical analysis; video coding; H.264 AVC; bitrate based adjustment scheme; flexible block size based mode decision; low-complexity mode decision; p-frame encoding; statistical data; video coding standards; Automatic voltage control; Bit rate; Communication standards; Degradation; Encoding; Lagrangian functions; MPEG 4 Standard; Motion compensation; Rate-distortion; Video coding; H.264/AVC; inter mode decision; macroblock;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Information, Communications & Signal Processing, 2007 6th International Conference on
  • Conference_Location
    Singapore
  • Print_ISBN
    978-1-4244-0982-2
  • Electronic_ISBN
    978-1-4244-0983-9
  • Type

    conf

  • DOI
    10.1109/ICICS.2007.4449778
  • Filename
    4449778