CHLAC特徴と Grid コンピューティングを併用したリアルタイム動作認識  [in Japanese] Realtime Motion Recognition by using CHLAC feature and Grid Computing  [in Japanese]

Search this Article

Author(s)

Abstract

CHLAC特徴は,動作認識においてその高い認識能力が報告されており,かつ比較的計算コストの低い特徴である.しかし,特徴数の増加に伴い実時間処理が困難になるという問題があった.そこで,立体高次局所自己相関(CHLAC)特徴を用いたカメラ1台による実時間動作認識システムの構築に関する研究を行った.本稿では,グリッド計算機群を用いた並列化手法を導入し,上記の課題を克服した.本システムを,格闘動作という素早い全身動作に対して適用した結果,実時間処理と高い認識性能の両立を達成した.

Cubic Higher-order Local AutoCorrelation (CHLAC) features are known for its good ability to recognize motions in spite of its relatively small computational cost. However, increase of the number of features prevents realtime processing. We studied on construction of a realtime motion recognition system with one camera using CHLAC features. To overcome this problem, we invented a system which combined a traditional CHLAC method and parallelizing techniques developed for grid computers. The system was applied to quick, whole-body fighting motions and successfully recognized them in realtime.

Journal

  • IEICE technical report

    IEICE technical report 105(615), 97-102, 2006-02-24

    The Institute of Electronics, Information and Communication Engineers

References:  12

Cited by:  11

Codes

  • NII Article ID (NAID)
    110004662983
  • NII NACSIS-CAT ID (NCID)
    AN10541106
  • Text Lang
    JPN
  • Article Type
    Journal Article
  • ISSN
    09135685
  • NDL Article ID
    7849779
  • NDL Source Classification
    ZN33(科学技術--電気工学・電気機械工業--電子工学・電気通信)
  • NDL Call No.
    Z16-940
  • Data Source
    CJP  CJPref  NDL  NII-ELS 
Page Top