Multi-view Multi-modal Gait Based Human Identity Recognition from Surveillance videos

Research output: A Conference proceeding or a Chapter in BookConference contribution

5 Citations (Scopus)

Abstract

In this paper we propose a novel human-identification scheme from long range gait profiles in surveillance videos. We investigate the role of multi view gait images acquired from multiple cameras, the importance of infrared and visible range images in ascertaining identity, the impact of multimodal fusion, efficient subspace features and classifier methods, and the role of soft/secondary biometric (walking style) in enhancing the accuracy and robustness of the identification systems, Experimental evaluation of several subspace based gait feature extraction approaches (PCA/LDA) and learning classifier methods (NB/MLP/SVM/SMO) on different datasets from a publicly available gait database CASIA, show significant improvement in recognition accuracies with multimodal fusion of multi-view gait images from visible and infrared cameras acquired from video surveillance scenarios.
Original languageEnglish
Title of host publicationMultimodal Pattern Recognition of Social Signals in Human-Computer-Interaction
EditorsFriedhelm Schwenker, Stefan Scherer, Louis-Philippe Morency
Place of PublicationBerling Heidelberg
PublisherSpringer
Pages88-99
Number of pages12
Volume7742
ISBN (Electronic)9783642370816
ISBN (Print)9783642370809
DOIs
Publication statusPublished - 2013
EventIAPR TC3 Workshop, MPRSS 2012, Revised Selected Papers: Multimodal Pattern Recognition of Social Signals in Human-Computer-Interaction - Tsukuba, Tsukuba, Japan
Duration: 11 Nov 201215 Nov 2012
http://www.icpr2012.org/ (Conference webpage)

Conference

ConferenceIAPR TC3 Workshop, MPRSS 2012, Revised Selected Papers
CountryJapan
CityTsukuba
Period11/11/1215/11/12
Internet address

Fingerprint

Classifiers
Fusion reactions
Cameras
Infrared radiation
Biometrics
Robustness (control systems)
Feature extraction
Identification (control systems)

Cite this

Hossain, E., CHETTY, G., & GOECKE, R. (2013). Multi-view Multi-modal Gait Based Human Identity Recognition from Surveillance videos. In F. Schwenker, S. Scherer, & L-P. Morency (Eds.), Multimodal Pattern Recognition of Social Signals in Human-Computer-Interaction (Vol. 7742, pp. 88-99). Berling Heidelberg: Springer. https://doi.org/10.1007/978-3-642-37081-6_11
Hossain, Emdad ; CHETTY, Girija ; GOECKE, Roland. / Multi-view Multi-modal Gait Based Human Identity Recognition from Surveillance videos. Multimodal Pattern Recognition of Social Signals in Human-Computer-Interaction. editor / Friedhelm Schwenker ; Stefan Scherer ; Louis-Philippe Morency. Vol. 7742 Berling Heidelberg : Springer, 2013. pp. 88-99
@inproceedings{0855e8a85e814a31af2eb508c829b9d7,
title = "Multi-view Multi-modal Gait Based Human Identity Recognition from Surveillance videos",
abstract = "In this paper we propose a novel human-identification scheme from long range gait profiles in surveillance videos. We investigate the role of multi view gait images acquired from multiple cameras, the importance of infrared and visible range images in ascertaining identity, the impact of multimodal fusion, efficient subspace features and classifier methods, and the role of soft/secondary biometric (walking style) in enhancing the accuracy and robustness of the identification systems, Experimental evaluation of several subspace based gait feature extraction approaches (PCA/LDA) and learning classifier methods (NB/MLP/SVM/SMO) on different datasets from a publicly available gait database CASIA, show significant improvement in recognition accuracies with multimodal fusion of multi-view gait images from visible and infrared cameras acquired from video surveillance scenarios.",
keywords = "Multiview, Identification, Fusion",
author = "Emdad Hossain and Girija CHETTY and Roland GOECKE",
year = "2013",
doi = "10.1007/978-3-642-37081-6_11",
language = "English",
isbn = "9783642370809",
volume = "7742",
pages = "88--99",
editor = "Friedhelm Schwenker and Stefan Scherer and Louis-Philippe Morency",
booktitle = "Multimodal Pattern Recognition of Social Signals in Human-Computer-Interaction",
publisher = "Springer",
address = "Netherlands",

}

Hossain, E, CHETTY, G & GOECKE, R 2013, Multi-view Multi-modal Gait Based Human Identity Recognition from Surveillance videos. in F Schwenker, S Scherer & L-P Morency (eds), Multimodal Pattern Recognition of Social Signals in Human-Computer-Interaction. vol. 7742, Springer, Berling Heidelberg, pp. 88-99, IAPR TC3 Workshop, MPRSS 2012, Revised Selected Papers, Tsukuba, Japan, 11/11/12. https://doi.org/10.1007/978-3-642-37081-6_11

Multi-view Multi-modal Gait Based Human Identity Recognition from Surveillance videos. / Hossain, Emdad; CHETTY, Girija; GOECKE, Roland.

Multimodal Pattern Recognition of Social Signals in Human-Computer-Interaction. ed. / Friedhelm Schwenker; Stefan Scherer; Louis-Philippe Morency. Vol. 7742 Berling Heidelberg : Springer, 2013. p. 88-99.

Research output: A Conference proceeding or a Chapter in BookConference contribution

TY - GEN

T1 - Multi-view Multi-modal Gait Based Human Identity Recognition from Surveillance videos

AU - Hossain, Emdad

AU - CHETTY, Girija

AU - GOECKE, Roland

PY - 2013

Y1 - 2013

N2 - In this paper we propose a novel human-identification scheme from long range gait profiles in surveillance videos. We investigate the role of multi view gait images acquired from multiple cameras, the importance of infrared and visible range images in ascertaining identity, the impact of multimodal fusion, efficient subspace features and classifier methods, and the role of soft/secondary biometric (walking style) in enhancing the accuracy and robustness of the identification systems, Experimental evaluation of several subspace based gait feature extraction approaches (PCA/LDA) and learning classifier methods (NB/MLP/SVM/SMO) on different datasets from a publicly available gait database CASIA, show significant improvement in recognition accuracies with multimodal fusion of multi-view gait images from visible and infrared cameras acquired from video surveillance scenarios.

AB - In this paper we propose a novel human-identification scheme from long range gait profiles in surveillance videos. We investigate the role of multi view gait images acquired from multiple cameras, the importance of infrared and visible range images in ascertaining identity, the impact of multimodal fusion, efficient subspace features and classifier methods, and the role of soft/secondary biometric (walking style) in enhancing the accuracy and robustness of the identification systems, Experimental evaluation of several subspace based gait feature extraction approaches (PCA/LDA) and learning classifier methods (NB/MLP/SVM/SMO) on different datasets from a publicly available gait database CASIA, show significant improvement in recognition accuracies with multimodal fusion of multi-view gait images from visible and infrared cameras acquired from video surveillance scenarios.

KW - Multiview

KW - Identification

KW - Fusion

U2 - 10.1007/978-3-642-37081-6_11

DO - 10.1007/978-3-642-37081-6_11

M3 - Conference contribution

SN - 9783642370809

VL - 7742

SP - 88

EP - 99

BT - Multimodal Pattern Recognition of Social Signals in Human-Computer-Interaction

A2 - Schwenker, Friedhelm

A2 - Scherer, Stefan

A2 - Morency, Louis-Philippe

PB - Springer

CY - Berling Heidelberg

ER -

Hossain E, CHETTY G, GOECKE R. Multi-view Multi-modal Gait Based Human Identity Recognition from Surveillance videos. In Schwenker F, Scherer S, Morency L-P, editors, Multimodal Pattern Recognition of Social Signals in Human-Computer-Interaction. Vol. 7742. Berling Heidelberg: Springer. 2013. p. 88-99 https://doi.org/10.1007/978-3-642-37081-6_11