Rule-based trajectory segmentation for modeling hand motion trajectory

Jounghoon Beh, David Han, Hanseok Ko

Research output: Contribution to journalArticle

22 Citations (Scopus)

Abstract

In this paper, we propose a simple but effective method of modeling hand gestures based on the angles and angular change rates of the hand trajectories. Each hand motion trajectory is composed of a unique series of straight and curved segments. In our Hidden Markov Model (HMM) implementation, these trajectories are modeled as a connected series of states analogous to the series of phonemes in speech recognition. The novelty of the work presented herein is that it provides an automated process of segmenting gesture trajectories based on a simple set of threshold values in the angular change measure. In order to represent the angular distribution of each separated state, the von Mises distribution is used. A likelihood based state segmentation was implemented in addition to the threshold based method to ensure that the gesture sets are segmented consistently. The proposed method can separate each angular state of the training data at the initialization step, thus providing a solution to mitigate the ambiguities on initializing the HMM. The effectiveness of the proposed method was demonstrated by the higher recognition rates in the experiments compared to the conventional methods.

Original languageEnglish
Pages (from-to)1586-1601
Number of pages16
JournalPattern Recognition
Volume47
Issue number4
DOIs
Publication statusPublished - 2014 Apr 1

Fingerprint

Trajectories
Hidden Markov models
Angular distribution
Speech recognition
Experiments

Keywords

  • Hand gesture recognition
  • Hidden Markov model
  • HMM initialization
  • Trajectory segmentation

ASJC Scopus subject areas

  • Software
  • Artificial Intelligence
  • Computer Vision and Pattern Recognition
  • Signal Processing

Cite this

Rule-based trajectory segmentation for modeling hand motion trajectory. / Beh, Jounghoon; Han, David; Ko, Hanseok.

In: Pattern Recognition, Vol. 47, No. 4, 01.04.2014, p. 1586-1601.

Research output: Contribution to journalArticle

Beh, Jounghoon ; Han, David ; Ko, Hanseok. / Rule-based trajectory segmentation for modeling hand motion trajectory. In: Pattern Recognition. 2014 ; Vol. 47, No. 4. pp. 1586-1601.
@article{8049607b59c5424da60df912d23d706b,
title = "Rule-based trajectory segmentation for modeling hand motion trajectory",
abstract = "In this paper, we propose a simple but effective method of modeling hand gestures based on the angles and angular change rates of the hand trajectories. Each hand motion trajectory is composed of a unique series of straight and curved segments. In our Hidden Markov Model (HMM) implementation, these trajectories are modeled as a connected series of states analogous to the series of phonemes in speech recognition. The novelty of the work presented herein is that it provides an automated process of segmenting gesture trajectories based on a simple set of threshold values in the angular change measure. In order to represent the angular distribution of each separated state, the von Mises distribution is used. A likelihood based state segmentation was implemented in addition to the threshold based method to ensure that the gesture sets are segmented consistently. The proposed method can separate each angular state of the training data at the initialization step, thus providing a solution to mitigate the ambiguities on initializing the HMM. The effectiveness of the proposed method was demonstrated by the higher recognition rates in the experiments compared to the conventional methods.",
keywords = "Hand gesture recognition, Hidden Markov model, HMM initialization, Trajectory segmentation",
author = "Jounghoon Beh and David Han and Hanseok Ko",
year = "2014",
month = "4",
day = "1",
doi = "10.1016/j.patcog.2013.11.010",
language = "English",
volume = "47",
pages = "1586--1601",
journal = "Pattern Recognition",
issn = "0031-3203",
publisher = "Elsevier Limited",
number = "4",

}

TY - JOUR

T1 - Rule-based trajectory segmentation for modeling hand motion trajectory

AU - Beh, Jounghoon

AU - Han, David

AU - Ko, Hanseok

PY - 2014/4/1

Y1 - 2014/4/1

N2 - In this paper, we propose a simple but effective method of modeling hand gestures based on the angles and angular change rates of the hand trajectories. Each hand motion trajectory is composed of a unique series of straight and curved segments. In our Hidden Markov Model (HMM) implementation, these trajectories are modeled as a connected series of states analogous to the series of phonemes in speech recognition. The novelty of the work presented herein is that it provides an automated process of segmenting gesture trajectories based on a simple set of threshold values in the angular change measure. In order to represent the angular distribution of each separated state, the von Mises distribution is used. A likelihood based state segmentation was implemented in addition to the threshold based method to ensure that the gesture sets are segmented consistently. The proposed method can separate each angular state of the training data at the initialization step, thus providing a solution to mitigate the ambiguities on initializing the HMM. The effectiveness of the proposed method was demonstrated by the higher recognition rates in the experiments compared to the conventional methods.

AB - In this paper, we propose a simple but effective method of modeling hand gestures based on the angles and angular change rates of the hand trajectories. Each hand motion trajectory is composed of a unique series of straight and curved segments. In our Hidden Markov Model (HMM) implementation, these trajectories are modeled as a connected series of states analogous to the series of phonemes in speech recognition. The novelty of the work presented herein is that it provides an automated process of segmenting gesture trajectories based on a simple set of threshold values in the angular change measure. In order to represent the angular distribution of each separated state, the von Mises distribution is used. A likelihood based state segmentation was implemented in addition to the threshold based method to ensure that the gesture sets are segmented consistently. The proposed method can separate each angular state of the training data at the initialization step, thus providing a solution to mitigate the ambiguities on initializing the HMM. The effectiveness of the proposed method was demonstrated by the higher recognition rates in the experiments compared to the conventional methods.

KW - Hand gesture recognition

KW - Hidden Markov model

KW - HMM initialization

KW - Trajectory segmentation

UR - http://www.scopus.com/inward/record.url?scp=84891626106&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84891626106&partnerID=8YFLogxK

U2 - 10.1016/j.patcog.2013.11.010

DO - 10.1016/j.patcog.2013.11.010

M3 - Article

AN - SCOPUS:84891626106

VL - 47

SP - 1586

EP - 1601

JO - Pattern Recognition

JF - Pattern Recognition

SN - 0031-3203

IS - 4

ER -