Human Motion Identification using the technique of Dynamic Time Warping

2 Citations 7 References
  • OSAKI Ryuta
    Department of Computer and Systems Engineering, Faculty of Engineering, Kobe University
  • UEHARA Kuniaki
    Research Center for Urban Safety and Security, Kobe University

Bibliographic Information

Other Title
  • Dynamic Time Warping 法を用いた身体運動の動作識別

Search this article

Description

A machine that can recognize human motions is available in a wide range of useful applications. However, recognition of human motion patterns is very difficult, even if the same person do the same motion;since speed, course or both of them can vary significantly each time. In this paper, we use the Dynamic Time Warping(DTW) technique to make a reasonable motion pattern transformation so as to regularize time variation which human motion patterns have. The machine uses the transformed motion pattern and recognizes it by a pattern matching approach. Furthermore, to reduce recognition time, we decrease the number of candidates, which are compared to transformed pattern, by using Fourier spectrum.

Journal

  • IPSJ SIG Notes

    IPSJ SIG Notes 116 (2), 233-240, 1998-07-09

    Information Processing Society of Japan (IPSJ)

Citations (2)*help

See more

References(7)*help

See more

Details 詳細情報について

  • CRID
    1573950401967072512
  • NII Article ID
    110002930847
  • NII Book ID
    AN10112482
  • ISSN
    09196072
  • Text Lang
    ja
  • Data Source
    • CiNii Articles

Report a problem

Back to top