Person identification from action styles

Igor Kviatkovsky; Ilan Shimshoni; Ehud Rivlin

doi:10.1109/CVPRW.2015.7301323

Person identification from action styles

Igor Kviatkovsky, Ilan Shimshoni, Ehud Rivlin

Department of Information Systems

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

We consider a problem of identifying people based on their styles in performing actions from an arbitrary predefined set of action types. We present a generative model describing the action instance creation process and derive a probabilistic identity inference scheme, which implicitly includes action type inference as one of its components. Our experiments validate the power of the approach. We report high recognition rates on four publicly available action recognition datasets and one dataset for person authentication, on which we obtain state-of-the-art results. We make use of existing action representations and show that combining them with an action-specific Mahalanobis metric, learned from examples, improves the results.

Original language	English
Title of host publication	2015 IEEE Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2015
Publisher	IEEE Computer Society
Pages	84-92
Number of pages	9
ISBN (Electronic)	9781467367592
DOIs	https://doi.org/10.1109/CVPRW.2015.7301323
State	Published - 19 Oct 2015
Event	IEEE Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2015 - Boston, United States Duration: 7 Jun 2015 → 12 Jun 2015

Publication series

Name	IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops
Volume	2015-October
ISSN (Print)	2160-7508
ISSN (Electronic)	2160-7516

Conference

Conference	IEEE Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2015
Country/Territory	United States
City	Boston
Period	7/06/15 → 12/06/15

Bibliographical note

Publisher Copyright:
© 2015 IEEE.

Keywords

Hidden Markov models
Joints
Measurement
Principal component analysis
Three-dimensional displays
Training

ASJC Scopus subject areas

Computer Vision and Pattern Recognition
Electrical and Electronic Engineering

Access to Document

10.1109/CVPRW.2015.7301323

Cite this

Kviatkovsky, I., Shimshoni, I., & Rivlin, E. (2015). Person identification from action styles. In 2015 IEEE Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2015 (pp. 84-92). Article 7301323 (IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops; Vol. 2015-October). IEEE Computer Society. https://doi.org/10.1109/CVPRW.2015.7301323

@inproceedings{5dc8289d9f62427b8b9798dec55aa11d,

title = "Person identification from action styles",

abstract = "We consider a problem of identifying people based on their styles in performing actions from an arbitrary predefined set of action types. We present a generative model describing the action instance creation process and derive a probabilistic identity inference scheme, which implicitly includes action type inference as one of its components. Our experiments validate the power of the approach. We report high recognition rates on four publicly available action recognition datasets and one dataset for person authentication, on which we obtain state-of-the-art results. We make use of existing action representations and show that combining them with an action-specific Mahalanobis metric, learned from examples, improves the results.",

keywords = "Hidden Markov models, Joints, Measurement, Principal component analysis, Three-dimensional displays, Training",

author = "Igor Kviatkovsky and Ilan Shimshoni and Ehud Rivlin",

note = "Publisher Copyright: {\textcopyright} 2015 IEEE.; IEEE Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2015 ; Conference date: 07-06-2015 Through 12-06-2015",

year = "2015",

month = oct,

day = "19",

doi = "10.1109/CVPRW.2015.7301323",

language = "English",

series = "IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops",

publisher = "IEEE Computer Society",

pages = "84--92",

booktitle = "2015 IEEE Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2015",

address = "United States",

}

Kviatkovsky, I, Shimshoni, I & Rivlin, E 2015, Person identification from action styles. in 2015 IEEE Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2015., 7301323, IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, vol. 2015-October, IEEE Computer Society, pp. 84-92, IEEE Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2015, Boston, United States, 7/06/15. https://doi.org/10.1109/CVPRW.2015.7301323

Person identification from action styles. / Kviatkovsky, Igor; Shimshoni, Ilan; Rivlin, Ehud.
2015 IEEE Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2015. IEEE Computer Society, 2015. p. 84-92 7301323 (IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops; Vol. 2015-October).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Person identification from action styles

AU - Kviatkovsky, Igor

AU - Shimshoni, Ilan

AU - Rivlin, Ehud

PY - 2015/10/19

Y1 - 2015/10/19

N2 - We consider a problem of identifying people based on their styles in performing actions from an arbitrary predefined set of action types. We present a generative model describing the action instance creation process and derive a probabilistic identity inference scheme, which implicitly includes action type inference as one of its components. Our experiments validate the power of the approach. We report high recognition rates on four publicly available action recognition datasets and one dataset for person authentication, on which we obtain state-of-the-art results. We make use of existing action representations and show that combining them with an action-specific Mahalanobis metric, learned from examples, improves the results.

AB - We consider a problem of identifying people based on their styles in performing actions from an arbitrary predefined set of action types. We present a generative model describing the action instance creation process and derive a probabilistic identity inference scheme, which implicitly includes action type inference as one of its components. Our experiments validate the power of the approach. We report high recognition rates on four publicly available action recognition datasets and one dataset for person authentication, on which we obtain state-of-the-art results. We make use of existing action representations and show that combining them with an action-specific Mahalanobis metric, learned from examples, improves the results.

KW - Hidden Markov models

KW - Joints

KW - Measurement

KW - Principal component analysis

KW - Three-dimensional displays

KW - Training

UR - http://www.scopus.com/inward/record.url?scp=84951957917&partnerID=8YFLogxK

U2 - 10.1109/CVPRW.2015.7301323

DO - 10.1109/CVPRW.2015.7301323

M3 - Conference contribution

AN - SCOPUS:84951957917

T3 - IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops

SP - 84

EP - 92

BT - 2015 IEEE Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2015

PB - IEEE Computer Society

T2 - IEEE Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2015

Y2 - 7 June 2015 through 12 June 2015

ER -

Person identification from action styles

Abstract

Publication series

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this