• DocumentCode
    2175505
  • Title

    Design of Dynamically Reconfigurable Processor for the H.264/AVC Image Prediction and De-blocking Filter

  • Author

    Hayakawa, Y. ; Kanasugi, A.

  • Author_Institution
    Grad. Sch. of Eng., Tokyo Denki Univ., Tokyo, Japan
  • fYear
    2010
  • fDate
    11-13 Dec. 2010
  • Firstpage
    127
  • Lastpage
    131
  • Abstract
    H.264/AVC provides high video quality at substantially low bit rates. However, the computational complexity of H.264/AVC is very high. A high-speed general-purpose processor is necessary to process H.264/AVC. However, it is difficult to use such a processor for a portable device. Therefore, an application-specific processor is necessary. A dynamic reconfiguration can virtually expand the circuit area in a limited chip area. Therefore, this article proposes a dynamically reconfigurable processor for H.264/AVC image prediction. H.264/AVC contains inter-prediction processes and de-blocking filter. The inter-prediction processes and de-blocking filter are not used at the same time. The proposed processor was designed and synthesized, and dynamically reconfigures those circuits. As a result, look up tables (LUTs) were reduced 10%, flip-flops were about the same, and the maximum delay was increased 10%.
  • Keywords
    computational complexity; flip-flops; image resolution; parallel processing; reconfigurable architectures; table lookup; video coding; H.264-AVC Image Prediction; chip area; circuit area; computational complexity; de-blocking filter; dynamically reconfigurable processor; flip-flops; high-speed general- purpose processor; look up tables; video quality; Adders; Automatic voltage control; Discrete cosine transforms; Encoding; Interpolation; Quantization; Table lookup; H.264/AVC; de-blocking filter; dynamically reconfiguration; inter prediction;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computational Science and Engineering (CSE), 2010 IEEE 13th International Conference on
  • Conference_Location
    Hong Kong
  • Print_ISBN
    978-1-4244-9591-7
  • Electronic_ISBN
    978-0-7695-4323-9
  • Type

    conf

  • DOI
    10.1109/CSE.2010.24
  • Filename
    5692466