Reconstructing and analyzing periodic human motion from stationary monocular views

Evan Ribnick, Ravishankar Sivalingam, Nikolaos P Papanikolopoulos, Kostas Daniilidis

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

We have shown previously that it is possible to accurately reconstruct periodic motions in 3D from a single camera view, using periodicity as a physical constraint from which to perform geometric inference. In this paper we explore the suitability of the reconstruction techniques for real human motion. We examine the degree of periodicity of human gait empirically, and develop algorithmic tools to address some of the challenges arising from this type of motion, including reconstructing motions that deviate from pure periodicity, properly handling the trajectories of multiple points on an articulated body, and proposing a distance function for measuring the difference between two reconstructions. Importantly, we illustrate the usefulness of these techniques by applying them to the tasks of view-invariant activity classification, clinical gait analysis and person identification.

Original languageEnglish (US)
Pages (from-to)815-826
Number of pages12
JournalComputer Vision and Image Understanding
Volume116
Issue number7
DOIs
StatePublished - Jul 1 2012

Fingerprint

Gait analysis
Cameras
Trajectories

Keywords

  • 3D reconstruction
  • Activity classification
  • Gait analysis
  • Human motion
  • Periodicity

Cite this

Reconstructing and analyzing periodic human motion from stationary monocular views. / Ribnick, Evan; Sivalingam, Ravishankar; Papanikolopoulos, Nikolaos P; Daniilidis, Kostas.

In: Computer Vision and Image Understanding, Vol. 116, No. 7, 01.07.2012, p. 815-826.

Research output: Contribution to journalArticle

Ribnick, Evan ; Sivalingam, Ravishankar ; Papanikolopoulos, Nikolaos P ; Daniilidis, Kostas. / Reconstructing and analyzing periodic human motion from stationary monocular views. In: Computer Vision and Image Understanding. 2012 ; Vol. 116, No. 7. pp. 815-826.
@article{7ead1d70874e4beb8a848c3467af1adc,
title = "Reconstructing and analyzing periodic human motion from stationary monocular views",
abstract = "We have shown previously that it is possible to accurately reconstruct periodic motions in 3D from a single camera view, using periodicity as a physical constraint from which to perform geometric inference. In this paper we explore the suitability of the reconstruction techniques for real human motion. We examine the degree of periodicity of human gait empirically, and develop algorithmic tools to address some of the challenges arising from this type of motion, including reconstructing motions that deviate from pure periodicity, properly handling the trajectories of multiple points on an articulated body, and proposing a distance function for measuring the difference between two reconstructions. Importantly, we illustrate the usefulness of these techniques by applying them to the tasks of view-invariant activity classification, clinical gait analysis and person identification.",
keywords = "3D reconstruction, Activity classification, Gait analysis, Human motion, Periodicity",
author = "Evan Ribnick and Ravishankar Sivalingam and Papanikolopoulos, {Nikolaos P} and Kostas Daniilidis",
year = "2012",
month = "7",
day = "1",
doi = "10.1016/j.cviu.2012.03.004",
language = "English (US)",
volume = "116",
pages = "815--826",
journal = "Computer Vision and Image Understanding",
issn = "1077-3142",
publisher = "Academic Press Inc.",
number = "7",

}

TY - JOUR

T1 - Reconstructing and analyzing periodic human motion from stationary monocular views

AU - Ribnick, Evan

AU - Sivalingam, Ravishankar

AU - Papanikolopoulos, Nikolaos P

AU - Daniilidis, Kostas

PY - 2012/7/1

Y1 - 2012/7/1

N2 - We have shown previously that it is possible to accurately reconstruct periodic motions in 3D from a single camera view, using periodicity as a physical constraint from which to perform geometric inference. In this paper we explore the suitability of the reconstruction techniques for real human motion. We examine the degree of periodicity of human gait empirically, and develop algorithmic tools to address some of the challenges arising from this type of motion, including reconstructing motions that deviate from pure periodicity, properly handling the trajectories of multiple points on an articulated body, and proposing a distance function for measuring the difference between two reconstructions. Importantly, we illustrate the usefulness of these techniques by applying them to the tasks of view-invariant activity classification, clinical gait analysis and person identification.

AB - We have shown previously that it is possible to accurately reconstruct periodic motions in 3D from a single camera view, using periodicity as a physical constraint from which to perform geometric inference. In this paper we explore the suitability of the reconstruction techniques for real human motion. We examine the degree of periodicity of human gait empirically, and develop algorithmic tools to address some of the challenges arising from this type of motion, including reconstructing motions that deviate from pure periodicity, properly handling the trajectories of multiple points on an articulated body, and proposing a distance function for measuring the difference between two reconstructions. Importantly, we illustrate the usefulness of these techniques by applying them to the tasks of view-invariant activity classification, clinical gait analysis and person identification.

KW - 3D reconstruction

KW - Activity classification

KW - Gait analysis

KW - Human motion

KW - Periodicity

UR - http://www.scopus.com/inward/record.url?scp=84859762672&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84859762672&partnerID=8YFLogxK

U2 - 10.1016/j.cviu.2012.03.004

DO - 10.1016/j.cviu.2012.03.004

M3 - Article

VL - 116

SP - 815

EP - 826

JO - Computer Vision and Image Understanding

JF - Computer Vision and Image Understanding

SN - 1077-3142

IS - 7

ER -