제스처 인식을 위한 은닉 마르코프 모델

Hidden Markov Model for Gesture Recognition

  • 박혜선 (경북대학교 컴퓨터공학과) ;
  • 김은이 (건국대학교 인터넷미디어학부) ;
  • 김항준 (경북대학교 컴퓨터공학과)
  • Park, Hye-Sun (Dept. of Computer Eng., Kyungpook National Univ.) ;
  • Kim, Eun-Yi (Dept. of Internet and Multimedia Engineering, NITRI, Konkuk Univ.) ;
  • Kim, Hang-Joon (Dept. of Computer Eng., Kyungpook National Univ.)
  • 발행 : 2006.01.01

초록

본 논문에서는 은닉 마르코프 모델 (HMM: hidden Markov model)을 이용한 제스처 인식 방법을 제안하고, 이를 게임 시스템의 인터페이스로 적용한 사례를 소개한다. 제안된 방법은 다음의 두 가지 특징을 가진다. 첫 번째는 사전에 분할된 데이터 열을 입력으로 사용하는 기존의 방법과는 달리, 제안된 방법은 카메라로부터 입력되는 비디오 스트림을 HMM의 입력으로 사용한다는 것이다. 두 번째는 제안된 HMM은 제스처의 분할과 인식을 동시에 수행한다는 것이다. 제안된 방법에서 사용자의 제스처는 13개의 제스처들을 인식하는 13개의 specific-HMM들을 결합하는 하나의 통합된 HMM을 통해 인식된다. 제안된 HMM은 사용자의 머리와 양손의 2D-위치 좌표로 구성된 포즈 심볼들의 열을 입력받는다. 그리고 새로운 포즈가 입력될 때마다, HMM의 상태 확률 값을 갱신한다. 그때, 만약 특정 상태의 확률 값이 미리 정해둔 임계치보다 큰 경우, 그 특정 상태를 포함하고 있는 제스처로 인식한다 제안된 방법의 정당성을 입증하기 위하여, 제안된 방법은 Quake II라는 컴퓨터 게임에 적용되었다. 실험결과는 제안된 방법이 높은 인식 정확률과, 계산 시간을 확연하게 감소시킬 수 있었음을 보여주었다.

This paper proposes a novel hidden Markov model (HMM)-based gesture recognition method and applies it to an HCI to control a computer game. The novelty of the proposed method is two-fold: 1) the proposed method uses a continuous streaming of human motion as the input to the HMM instead of isolated data sequences or pre-segmented sequences of data and 2) the gesture segmentation and recognition are performed simultaneously. The proposed method consists of a single HMM composed of thirteen gesture-specific HMMs that independently recognize certain gestures. It takes a continuous stream of pose symbols as an input, where a pose is composed of coordinates that indicate the face, left hand, and right hand. Whenever a new input Pose arrives, the HMM continuously updates its state probabilities, then recognizes a gesture if the probability of a distinctive state exceeds a predefined threshold. To assess the validity of the proposed method, it was applied to a real game, Quake II, and the results demonstrated that the proposed HMM could provide very useful information to enhance the discrimination between different classes and reduce the computational cost.

키워드

참고문헌

  1. T. Frantti, S. Kallio, Expert system for gesture recognition in terminal's user interface, Expert Systems with Applicaiions, Vol. 26, 189-202 (2004) https://doi.org/10.1016/S0957-4174(03)00134-9
  2. H. Kang, C.W. Lee, K lung, Recognition-based gesture spotting in video games, Pattern Recognition Letters, Vol. 25, 1701-1714, (2004) https://doi.org/10.1016/j.patrec.2004.06.016
  3. W.N. Chan, S. Ranganath, Real-time gesture recogrution system and application, Image and Vision Computing, Vol. 20, 993-1007, (2002) https://doi.org/10.1016/S0262-8856(02)00113-0
  4. P. Marco, Vision-based user interlaces: methods and applications, International Journal of Human-Computer Studies, Vol. 57, 27-73, (2002) https://doi.org/10.1006/ijhc.2002.1012
  5. K. Oka, Y. Sato, H Koike, Real-time tracking of multiple fingertips and gesture recognition for augmented desk interface systems, Automatic Face and Gesture Recognition, 411-416, (2002) https://doi.org/10.1109/AFGR.2002.1004191
  6. H.S. Yoon, J. Soh, Y.J. Bae, S.Y. Yang, Hand gesture recognition using combined features of location, angle and velocity, Pattern Recognition, Vol. 34, 1491-1501, (2001) https://doi.org/10.1016/S0031-3203(00)00096-0
  7. L. Gupta, S. Ma, Gesture-based interaction and communication: automated classification of hand gesture contours, IEEE Transactions on Systems, Man and Cybernetics, Part C, 114-120, (2001) https://doi.org/10.1109/5326.923274
  8. I. Cohen, N. Sebe, A. Garg, L.S. Chen, S.H Thomas, Facial expression recognition from video sequences: temporal and static modeling, Computer Vision and Image Understanding, Vol. 91, 160-187, (2003) https://doi.org/10.1016/S1077-3142(03)00081-X
  9. T. Otsuka, J. Ohya, An HMM-based approach for off-line unconstrained handwritten word modeling and recognition, IEEE Transaction on Pattern Analysis and Machine Intelligence, Vol. 21, 752-760, (1999) https://doi.org/10.1109/34.784288
  10. F. Quek, Toward a vision-based human gesture interface, Conf. on Virtual Reality Software Technol., 17-31, (1994)
  11. A. Kendon, Current issues in the study of gesture, The Biological Foundation of Gestures: Motor and Semiotic Aspects. Lawrence Erlbaum Associate, 23-47, (1986)
  12. H.K. Lee, J.H. Kim, An HMM-based threshold model approach for gesture recognition, IEEE Transaction on Pattern Analysis and Machine Intelligence, Vol. 21, 961-973, (1999) https://doi.org/10.1109/34.799904