본문 바로가기
  • Home

Recognition of conducting motion using HMM

  • Journal of The Korea Society of Computer and Information
  • Abbr : JKSCI
  • 2004, 9(1), pp.25-30
  • Publisher : The Korean Society Of Computer And Information
  • Research Area : Engineering > Computer Science

문형득 1 KOO JA YOUNG 1

1단국대학교

ABSTRACT

In this paper, a beat recognition method from a sequence of images of conducting person was proposed. Hand position was detected using color discrimination, and symbolized by quantization. Then a motion of the conductor was represented as a sequence of symbols. HMM (Hidden Markov Model), which is excellent for recognition of sequence pattern with some level of variation, was used to recognize the sequence of symbols to be a motion for a beat.

Citation status

* References for papers published after 2022 are currently being built.