Human gait recognition using localized Grassmann mean representatives with partial least squares regression

Tee Connie, Michael Kah Ong Goh, Andrew Beng Jin Teoh

Research output: Contribution to journalArticle

Abstract

Gait recognition has become popular due to the rising demand for nonintrusive biometrics. At its nascent stage of development, gait recognition faces a number of challenges. The performance of a gait recognition system is sensitive towards factors like viewing angle, clothing, shoe type, load carriage and speed changes. In this paper, the problems of gait are formulated on the Grassmann manifold. It is not difficult to obtain multiple snapshots of a walking subjects with the wide availability of camera networks. These sets of images can be modelled as low-dimensional subspaces, which can be realized naturally as points on the Grassmann manifold. Modelling image sets as low-dimensional subspaces provides not only possible clue of one’s gait, but also the common patterns of variation in the set. We present a method called Localized Grassmann Mean Representatives with Partial Least Squares Regression (LoGPLS) to infer a low-dimensional Euclidean approximation of the manifold. The notion of local mean representatives is introduced to construct multiple tangent spaces to better approximate the topological structure of the manifold. As the properties of the tangent spaces allows the Grassmann points to be evaluated in the vector space, partial least squares is applied to allow a more accurate classification of the points in a reduced space. Experiments have been conducted on four different publicly available gait databases. Empirical evidences demonstrate the effectiveness of the proposed approach in solving the various covariates in gait recognition.

Original languageEnglish
Pages (from-to)28457-28482
Number of pages26
JournalMultimedia Tools and Applications
Volume77
Issue number21
DOIs
Publication statusPublished - 2018 Nov 1

Fingerprint

Biometrics
Vector spaces
Face recognition
Cameras
Availability
Experiments

All Science Journal Classification (ASJC) codes

  • Software
  • Media Technology
  • Hardware and Architecture
  • Computer Networks and Communications

Cite this

@article{443ba4bd9e644f1596ef263005923042,
title = "Human gait recognition using localized Grassmann mean representatives with partial least squares regression",
abstract = "Gait recognition has become popular due to the rising demand for nonintrusive biometrics. At its nascent stage of development, gait recognition faces a number of challenges. The performance of a gait recognition system is sensitive towards factors like viewing angle, clothing, shoe type, load carriage and speed changes. In this paper, the problems of gait are formulated on the Grassmann manifold. It is not difficult to obtain multiple snapshots of a walking subjects with the wide availability of camera networks. These sets of images can be modelled as low-dimensional subspaces, which can be realized naturally as points on the Grassmann manifold. Modelling image sets as low-dimensional subspaces provides not only possible clue of one’s gait, but also the common patterns of variation in the set. We present a method called Localized Grassmann Mean Representatives with Partial Least Squares Regression (LoGPLS) to infer a low-dimensional Euclidean approximation of the manifold. The notion of local mean representatives is introduced to construct multiple tangent spaces to better approximate the topological structure of the manifold. As the properties of the tangent spaces allows the Grassmann points to be evaluated in the vector space, partial least squares is applied to allow a more accurate classification of the points in a reduced space. Experiments have been conducted on four different publicly available gait databases. Empirical evidences demonstrate the effectiveness of the proposed approach in solving the various covariates in gait recognition.",
author = "Tee Connie and Goh, {Michael Kah Ong} and Teoh, {Andrew Beng Jin}",
year = "2018",
month = "11",
day = "1",
doi = "10.1007/s11042-018-6045-y",
language = "English",
volume = "77",
pages = "28457--28482",
journal = "Multimedia Tools and Applications",
issn = "1380-7501",
publisher = "Springer Netherlands",
number = "21",

}

Human gait recognition using localized Grassmann mean representatives with partial least squares regression. / Connie, Tee; Goh, Michael Kah Ong; Teoh, Andrew Beng Jin.

In: Multimedia Tools and Applications, Vol. 77, No. 21, 01.11.2018, p. 28457-28482.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Human gait recognition using localized Grassmann mean representatives with partial least squares regression

AU - Connie, Tee

AU - Goh, Michael Kah Ong

AU - Teoh, Andrew Beng Jin

PY - 2018/11/1

Y1 - 2018/11/1

N2 - Gait recognition has become popular due to the rising demand for nonintrusive biometrics. At its nascent stage of development, gait recognition faces a number of challenges. The performance of a gait recognition system is sensitive towards factors like viewing angle, clothing, shoe type, load carriage and speed changes. In this paper, the problems of gait are formulated on the Grassmann manifold. It is not difficult to obtain multiple snapshots of a walking subjects with the wide availability of camera networks. These sets of images can be modelled as low-dimensional subspaces, which can be realized naturally as points on the Grassmann manifold. Modelling image sets as low-dimensional subspaces provides not only possible clue of one’s gait, but also the common patterns of variation in the set. We present a method called Localized Grassmann Mean Representatives with Partial Least Squares Regression (LoGPLS) to infer a low-dimensional Euclidean approximation of the manifold. The notion of local mean representatives is introduced to construct multiple tangent spaces to better approximate the topological structure of the manifold. As the properties of the tangent spaces allows the Grassmann points to be evaluated in the vector space, partial least squares is applied to allow a more accurate classification of the points in a reduced space. Experiments have been conducted on four different publicly available gait databases. Empirical evidences demonstrate the effectiveness of the proposed approach in solving the various covariates in gait recognition.

AB - Gait recognition has become popular due to the rising demand for nonintrusive biometrics. At its nascent stage of development, gait recognition faces a number of challenges. The performance of a gait recognition system is sensitive towards factors like viewing angle, clothing, shoe type, load carriage and speed changes. In this paper, the problems of gait are formulated on the Grassmann manifold. It is not difficult to obtain multiple snapshots of a walking subjects with the wide availability of camera networks. These sets of images can be modelled as low-dimensional subspaces, which can be realized naturally as points on the Grassmann manifold. Modelling image sets as low-dimensional subspaces provides not only possible clue of one’s gait, but also the common patterns of variation in the set. We present a method called Localized Grassmann Mean Representatives with Partial Least Squares Regression (LoGPLS) to infer a low-dimensional Euclidean approximation of the manifold. The notion of local mean representatives is introduced to construct multiple tangent spaces to better approximate the topological structure of the manifold. As the properties of the tangent spaces allows the Grassmann points to be evaluated in the vector space, partial least squares is applied to allow a more accurate classification of the points in a reduced space. Experiments have been conducted on four different publicly available gait databases. Empirical evidences demonstrate the effectiveness of the proposed approach in solving the various covariates in gait recognition.

UR - http://www.scopus.com/inward/record.url?scp=85046163748&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85046163748&partnerID=8YFLogxK

U2 - 10.1007/s11042-018-6045-y

DO - 10.1007/s11042-018-6045-y

M3 - Article

AN - SCOPUS:85046163748

VL - 77

SP - 28457

EP - 28482

JO - Multimedia Tools and Applications

JF - Multimedia Tools and Applications

SN - 1380-7501

IS - 21

ER -