Human identification based on a kinematical data of a gait

Adam Świtoński; Romualda Mucha; Dariusz Danowski; Monika Mucha; Grzegorz Cieślar; Konrad Wojciechowski; Aleksander Sieroń

Human identification based on a kinematical data of a gait

Adam Świtoński
, Romualda Mucha
, Dariusz Danowski
, Monika Mucha
, Grzegorz Cieślar
, Konrad Wojciechowski
, Aleksander Sieroń

Department of Graphics, Computer Vision and Digital Systems (RAU/RAU6)

Polish-Japanese Academy of Information Technology
Medical University of Silesia in Katowice
Goczalkowice Zdrój

Research output: Contribution to journal › Article › peer-review

6 Citations (Scopus)

Abstract

The paper is devoted to the gait identification challenges. It evaluates human abilities to recognize gait on the basis of skeleton animations. Further, it proposes the method of gait identification based on the kinematical data.The feature extraction approach and supervised learning are applied. To explore the most individual joints movements, aggregated feature rankings are calculated. To examine the proposed method, the database containing 353 gaits of 25 different actors is collected in the motion capture laboratory. We have obtained 99.7% of classification accuracy.

Translated title of the contribution	Identyfikacja osobnicza na podstawie kinematycznych danych chodu
Original language	English
Pages (from-to)	169-172
Number of pages	4
Journal	Przeglad Elektrotechniczny
Volume	87
Issue number	12 B
Publication status	Published - 2011

Keywords

Biometrics
Feature extraction
Feature selection
Gait identification
Motion capture
Supervised learning

ASJC Scopus subject areas

Electrical and Electronic Engineering

Cite this

@article{4a715b03daa54626896a5bb4c6abc186,

title = "Human identification based on a kinematical data of a gait",

abstract = "The paper is devoted to the gait identification challenges. It evaluates human abilities to recognize gait on the basis of skeleton animations. Further, it proposes the method of gait identification based on the kinematical data.The feature extraction approach and supervised learning are applied. To explore the most individual joints movements, aggregated feature rankings are calculated. To examine the proposed method, the database containing 353 gaits of 25 different actors is collected in the motion capture laboratory. We have obtained 99.7\% of classification accuracy.",

keywords = "Biometrics, Feature extraction, Feature selection, Gait identification, Motion capture, Supervised learning",

author = "Adam {\'S}wito{\'n}ski and Romualda Mucha and Dariusz Danowski and Monika Mucha and Grzegorz Cie{\'s}lar and Konrad Wojciechowski and Aleksander Siero{\'n}",

year = "2011",

language = "English",

volume = "87",

pages = "169--172",

journal = "Przeglad Elektrotechniczny",

issn = "0033-2097",

publisher = "Wydawnictwo SIGMA-NOT",

number = "12 B",

}

TY - JOUR

T1 - Human identification based on a kinematical data of a gait

AU - Świtoński, Adam

AU - Mucha, Romualda

AU - Danowski, Dariusz

AU - Mucha, Monika

AU - Cieślar, Grzegorz

AU - Wojciechowski, Konrad

AU - Sieroń, Aleksander

PY - 2011

Y1 - 2011

N2 - The paper is devoted to the gait identification challenges. It evaluates human abilities to recognize gait on the basis of skeleton animations. Further, it proposes the method of gait identification based on the kinematical data.The feature extraction approach and supervised learning are applied. To explore the most individual joints movements, aggregated feature rankings are calculated. To examine the proposed method, the database containing 353 gaits of 25 different actors is collected in the motion capture laboratory. We have obtained 99.7% of classification accuracy.

AB - The paper is devoted to the gait identification challenges. It evaluates human abilities to recognize gait on the basis of skeleton animations. Further, it proposes the method of gait identification based on the kinematical data.The feature extraction approach and supervised learning are applied. To explore the most individual joints movements, aggregated feature rankings are calculated. To examine the proposed method, the database containing 353 gaits of 25 different actors is collected in the motion capture laboratory. We have obtained 99.7% of classification accuracy.

KW - Biometrics

KW - Feature extraction

KW - Feature selection

KW - Gait identification

KW - Motion capture

KW - Supervised learning

UR - https://www.scopus.com/pages/publications/82655189511

M3 - Article

AN - SCOPUS:82655189511

SN - 0033-2097

VL - 87

SP - 169

EP - 172

JO - Przeglad Elektrotechniczny

JF - Przeglad Elektrotechniczny

IS - 12 B

ER -

Human identification based on a kinematical data of a gait

Abstract

Keywords

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this