Invariant Gait Continuum Based on the Duty-Factor

Preben Fihl, Thomas B. Moeslund

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

7 Citationer (Scopus)

Resumé

In this paper we present a method to describe the continuum of human gait in an invariant manner. The gait description is based on the duty-factor which is adopted from the biomechanics literature. We generate a database of artificial silhouettes representing the three main types of gait, i.e. walking, jogging, and running. By generating silhouettes from different camera angles we make the method invariant to camera viewpoint and to changing directions of movement. Silhouettes are extracted using the Code-book method and represented in a scale- and translation-invariant manner by using shape contexts and tangent orientations. Input silhouettes are matched to the database using the Hungarian method. We define a classifier based on the dissimilarity between the input silhouettes and the gait actions of the database. This classification achieves an overall recognition rate of 87.1% on a diverse test set, which is better than that achieved by other approaches applied to similar data. We extend this classification and results show that our representation of the gait continuum preserves the main features of the duty-factor.

OriginalsprogEngelsk
TidsskriftSignal, Image and Video Processing
Vol/bind3
Udgave nummer4
Sider (fra-til)391-402
Antal sider12
ISSN1863-1703
DOI
StatusUdgivet - 6 nov. 2008

Fingerprint

Cameras
Biomechanics
Classifiers

Citer dette

@article{3f0fb6c0951111dda004000ea68e967b,
title = "Invariant Gait Continuum Based on the Duty-Factor",
abstract = "In this paper we present a method to describe the continuum of human gait in an invariant manner. The gait description is based on the duty-factor which is adopted from the biomechanics literature. We generate a database of artificial silhouettes representing the three main types of gait, i.e. walking, jogging, and running. By generating silhouettes from different camera angles we make the method invariant to camera viewpoint and to changing directions of movement. Silhouettes are extracted using the Code-book method and represented in a scale- and translation-invariant manner by using shape contexts and tangent orientations. Input silhouettes are matched to the database using the Hungarian method. We define a classifier based on the dissimilarity between the input silhouettes and the gait actions of the database. This classification achieves an overall recognition rate of 87.1{\%} on a diverse test set, which is better than that achieved by other approaches applied to similar data. We extend this classification and results show that our representation of the gait continuum preserves the main features of the duty-factor.",
keywords = "Computer Vision, Human motion, Gait Analysis, Action recognition, Gait continuum",
author = "Preben Fihl and Moeslund, {Thomas B.}",
year = "2008",
month = "11",
day = "6",
doi = "10.1007/s11760-008-0089-9",
language = "English",
volume = "3",
pages = "391--402",
journal = "Signal, Image and Video Processing",
issn = "1863-1703",
publisher = "Springer",
number = "4",

}

Invariant Gait Continuum Based on the Duty-Factor. / Fihl, Preben; Moeslund, Thomas B.

I: Signal, Image and Video Processing, Bind 3, Nr. 4, 06.11.2008, s. 391-402.

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

TY - JOUR

T1 - Invariant Gait Continuum Based on the Duty-Factor

AU - Fihl, Preben

AU - Moeslund, Thomas B.

PY - 2008/11/6

Y1 - 2008/11/6

N2 - In this paper we present a method to describe the continuum of human gait in an invariant manner. The gait description is based on the duty-factor which is adopted from the biomechanics literature. We generate a database of artificial silhouettes representing the three main types of gait, i.e. walking, jogging, and running. By generating silhouettes from different camera angles we make the method invariant to camera viewpoint and to changing directions of movement. Silhouettes are extracted using the Code-book method and represented in a scale- and translation-invariant manner by using shape contexts and tangent orientations. Input silhouettes are matched to the database using the Hungarian method. We define a classifier based on the dissimilarity between the input silhouettes and the gait actions of the database. This classification achieves an overall recognition rate of 87.1% on a diverse test set, which is better than that achieved by other approaches applied to similar data. We extend this classification and results show that our representation of the gait continuum preserves the main features of the duty-factor.

AB - In this paper we present a method to describe the continuum of human gait in an invariant manner. The gait description is based on the duty-factor which is adopted from the biomechanics literature. We generate a database of artificial silhouettes representing the three main types of gait, i.e. walking, jogging, and running. By generating silhouettes from different camera angles we make the method invariant to camera viewpoint and to changing directions of movement. Silhouettes are extracted using the Code-book method and represented in a scale- and translation-invariant manner by using shape contexts and tangent orientations. Input silhouettes are matched to the database using the Hungarian method. We define a classifier based on the dissimilarity between the input silhouettes and the gait actions of the database. This classification achieves an overall recognition rate of 87.1% on a diverse test set, which is better than that achieved by other approaches applied to similar data. We extend this classification and results show that our representation of the gait continuum preserves the main features of the duty-factor.

KW - Computer Vision

KW - Human motion

KW - Gait Analysis

KW - Action recognition

KW - Gait continuum

U2 - 10.1007/s11760-008-0089-9

DO - 10.1007/s11760-008-0089-9

M3 - Journal article

VL - 3

SP - 391

EP - 402

JO - Signal, Image and Video Processing

JF - Signal, Image and Video Processing

SN - 1863-1703

IS - 4

ER -