A human motion estimation method based on GP-UKF

Ziyou Wang; Jun Kinugawa; Hongbo Wang; Kazuhiro Kosuge

doi:10.1109/ICInfA.2014.6932836

A human motion estimation method based on GP-UKF

Ziyou Wang, Jun Kinugawa, Hongbo Wang, Kazuhiro Kosuge

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

6 Citations (Scopus)

Abstract

A novel human motion estimation method is presented in this paper. The motion of the human is estimated by an Unscented Kalman filter (UKF), in which a nonlinear dynamic model is used to predict trajectory of human. This dynamic model is obtained from sample data by using Gaussian Process (GP) regression. The sample data includes information of body segment posture and trajectory data collected by motion capture system. The GP-UKF can extract the underlying dynamics from the sample data, with which the future non-linear transition can be predicted. The experiment results show that the proposed method has improved accuracy over conventional method.

Original language	English
Title of host publication	2014 IEEE International Conference on Information and Automation, ICIA 2014
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1228-1232
Number of pages	5
ISBN (Electronic)	9781479941001
DOIs	https://doi.org/10.1109/ICInfA.2014.6932836
Publication status	Published - 2014 Oct 21
Event	2014 IEEE International Conference on Information and Automation, ICIA 2014 - Hailar, Hulunbuir, China Duration: 2014 Jul 28 → 2014 Jul 30

Publication series

Name	2014 IEEE International Conference on Information and Automation, ICIA 2014

Other

Other	2014 IEEE International Conference on Information and Automation, ICIA 2014
Country/Territory	China
City	Hailar, Hulunbuir
Period	14/7/28 → 14/7/30

Keywords

GP-UKF
Gaussian Process
Motion estimation
Unscented Kalman filter

ASJC Scopus subject areas

Modelling and Simulation

Access to Document

10.1109/ICInfA.2014.6932836

Cite this

Wang, Z., Kinugawa, J., Wang, H., & Kosuge, K. (2014). A human motion estimation method based on GP-UKF. In 2014 IEEE International Conference on Information and Automation, ICIA 2014 (pp. 1228-1232). [6932836] (2014 IEEE International Conference on Information and Automation, ICIA 2014). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICInfA.2014.6932836

A human motion estimation method based on GP-UKF. / Wang, Ziyou; Kinugawa, Jun; Wang, Hongbo et al.

2014 IEEE International Conference on Information and Automation, ICIA 2014. Institute of Electrical and Electronics Engineers Inc., 2014. p. 1228-1232 6932836 (2014 IEEE International Conference on Information and Automation, ICIA 2014).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Wang, Z, Kinugawa, J, Wang, H & Kosuge, K 2014, A human motion estimation method based on GP-UKF. in 2014 IEEE International Conference on Information and Automation, ICIA 2014., 6932836, 2014 IEEE International Conference on Information and Automation, ICIA 2014, Institute of Electrical and Electronics Engineers Inc., pp. 1228-1232, 2014 IEEE International Conference on Information and Automation, ICIA 2014, Hailar, Hulunbuir, China, 14/7/28. https://doi.org/10.1109/ICInfA.2014.6932836

@inproceedings{a4d0c0dd2a694f7dbe03c58028540d42,

title = "A human motion estimation method based on GP-UKF",

abstract = "A novel human motion estimation method is presented in this paper. The motion of the human is estimated by an Unscented Kalman filter (UKF), in which a nonlinear dynamic model is used to predict trajectory of human. This dynamic model is obtained from sample data by using Gaussian Process (GP) regression. The sample data includes information of body segment posture and trajectory data collected by motion capture system. The GP-UKF can extract the underlying dynamics from the sample data, with which the future non-linear transition can be predicted. The experiment results show that the proposed method has improved accuracy over conventional method.",

keywords = "GP-UKF, Gaussian Process, Motion estimation, Unscented Kalman filter",

author = "Ziyou Wang and Jun Kinugawa and Hongbo Wang and Kazuhiro Kosuge",

note = "Publisher Copyright: {\textcopyright} 2014 IEEE.; 2014 IEEE International Conference on Information and Automation, ICIA 2014 ; Conference date: 28-07-2014 Through 30-07-2014",

year = "2014",

month = oct,

day = "21",

doi = "10.1109/ICInfA.2014.6932836",

language = "English",

series = "2014 IEEE International Conference on Information and Automation, ICIA 2014",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "1228--1232",

booktitle = "2014 IEEE International Conference on Information and Automation, ICIA 2014",

}

TY - GEN

T1 - A human motion estimation method based on GP-UKF

AU - Wang, Ziyou

AU - Kinugawa, Jun

AU - Wang, Hongbo

AU - Kosuge, Kazuhiro

PY - 2014/10/21

Y1 - 2014/10/21

N2 - A novel human motion estimation method is presented in this paper. The motion of the human is estimated by an Unscented Kalman filter (UKF), in which a nonlinear dynamic model is used to predict trajectory of human. This dynamic model is obtained from sample data by using Gaussian Process (GP) regression. The sample data includes information of body segment posture and trajectory data collected by motion capture system. The GP-UKF can extract the underlying dynamics from the sample data, with which the future non-linear transition can be predicted. The experiment results show that the proposed method has improved accuracy over conventional method.

AB - A novel human motion estimation method is presented in this paper. The motion of the human is estimated by an Unscented Kalman filter (UKF), in which a nonlinear dynamic model is used to predict trajectory of human. This dynamic model is obtained from sample data by using Gaussian Process (GP) regression. The sample data includes information of body segment posture and trajectory data collected by motion capture system. The GP-UKF can extract the underlying dynamics from the sample data, with which the future non-linear transition can be predicted. The experiment results show that the proposed method has improved accuracy over conventional method.

KW - GP-UKF

KW - Gaussian Process

KW - Motion estimation

KW - Unscented Kalman filter

UR - http://www.scopus.com/inward/record.url?scp=84914094518&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84914094518&partnerID=8YFLogxK

U2 - 10.1109/ICInfA.2014.6932836

DO - 10.1109/ICInfA.2014.6932836

M3 - Conference contribution

AN - SCOPUS:84914094518

T3 - 2014 IEEE International Conference on Information and Automation, ICIA 2014

SP - 1228

EP - 1232

BT - 2014 IEEE International Conference on Information and Automation, ICIA 2014

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2014 IEEE International Conference on Information and Automation, ICIA 2014

Y2 - 28 July 2014 through 30 July 2014

ER -

A human motion estimation method based on GP-UKF

Abstract

Publication series

Other

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this