Local inter-session variability modelling for object classification

Kaneswaran Anantharajah, Zong Yuan Ge, Chris McCool, Simon Denman, Clinton Fookes, Peter Corke, Dian Tjondronegoro, Sridha Sridharan

Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

15 Citations (Scopus)

Abstract

Object classification is plagued by the issue of session variation. Session variation describes any variation that makes one instance of an object look different to another, for instance due to pose or illumination variation. Recent work in the challenging task of face verification has shown that session variability modelling provides a mechanism to overcome some of these limitations. However, for computer vision purposes, it has only been applied in the limited setting of face verification. In this paper we propose a local region based intersession variability (ISV) modelling approach, and apply it to challenging real-world data. We propose a region based session variability modelling approach so that local session variations can be modelled, termed Local ISV. We then demonstrate the efficacy of this technique on a challenging real-world fish image database which includes images taken underwater, providing significant real-world session variations. This Local ISV approach provides a relative performance improvement of, on average, 23% on the challenging MOBIO, Multi-PIE and SCface face databases. It also provides a relative performance improvement of 35% on our challenging fish image dataset.

Original languageEnglish
Title of host publication2014 IEEE Winter Conference on Applications of Computer Vision, WACV 2014
EditorsWalter Scheirer, Ruigang Yang, Charles Stewart
Place of PublicationPiscataway NJ USA
PublisherIEEE, Institute of Electrical and Electronics Engineers
Pages309-316
Number of pages8
ISBN (Print)9781479949847
DOIs
Publication statusPublished - 1 Jan 2014
Externally publishedYes
EventIEEE Winter Conference on Applications of Computer Vision 2014 - Steamboat Springs, United States of America
Duration: 24 Mar 201426 Mar 2014
http://www.wacv14.org/

Conference

ConferenceIEEE Winter Conference on Applications of Computer Vision 2014
Abbreviated titleWACV 2014
CountryUnited States of America
CitySteamboat Springs
Period24/03/1426/03/14
Internet address

Cite this

Anantharajah, K., Ge, Z. Y., McCool, C., Denman, S., Fookes, C., Corke, P., ... Sridharan, S. (2014). Local inter-session variability modelling for object classification. In W. Scheirer, R. Yang, & C. Stewart (Eds.), 2014 IEEE Winter Conference on Applications of Computer Vision, WACV 2014 (pp. 309-316). [6836084] Piscataway NJ USA: IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/WACV.2014.6836084
Anantharajah, Kaneswaran ; Ge, Zong Yuan ; McCool, Chris ; Denman, Simon ; Fookes, Clinton ; Corke, Peter ; Tjondronegoro, Dian ; Sridharan, Sridha. / Local inter-session variability modelling for object classification. 2014 IEEE Winter Conference on Applications of Computer Vision, WACV 2014. editor / Walter Scheirer ; Ruigang Yang ; Charles Stewart. Piscataway NJ USA : IEEE, Institute of Electrical and Electronics Engineers, 2014. pp. 309-316
@inproceedings{3f85c87bf53c40b4bc59181dc041da1a,
title = "Local inter-session variability modelling for object classification",
abstract = "Object classification is plagued by the issue of session variation. Session variation describes any variation that makes one instance of an object look different to another, for instance due to pose or illumination variation. Recent work in the challenging task of face verification has shown that session variability modelling provides a mechanism to overcome some of these limitations. However, for computer vision purposes, it has only been applied in the limited setting of face verification. In this paper we propose a local region based intersession variability (ISV) modelling approach, and apply it to challenging real-world data. We propose a region based session variability modelling approach so that local session variations can be modelled, termed Local ISV. We then demonstrate the efficacy of this technique on a challenging real-world fish image database which includes images taken underwater, providing significant real-world session variations. This Local ISV approach provides a relative performance improvement of, on average, 23{\%} on the challenging MOBIO, Multi-PIE and SCface face databases. It also provides a relative performance improvement of 35{\%} on our challenging fish image dataset.",
author = "Kaneswaran Anantharajah and Ge, {Zong Yuan} and Chris McCool and Simon Denman and Clinton Fookes and Peter Corke and Dian Tjondronegoro and Sridha Sridharan",
year = "2014",
month = "1",
day = "1",
doi = "10.1109/WACV.2014.6836084",
language = "English",
isbn = "9781479949847",
pages = "309--316",
editor = "Walter Scheirer and Ruigang Yang and Charles Stewart",
booktitle = "2014 IEEE Winter Conference on Applications of Computer Vision, WACV 2014",
publisher = "IEEE, Institute of Electrical and Electronics Engineers",
address = "United States of America",

}

Anantharajah, K, Ge, ZY, McCool, C, Denman, S, Fookes, C, Corke, P, Tjondronegoro, D & Sridharan, S 2014, Local inter-session variability modelling for object classification. in W Scheirer, R Yang & C Stewart (eds), 2014 IEEE Winter Conference on Applications of Computer Vision, WACV 2014., 6836084, IEEE, Institute of Electrical and Electronics Engineers, Piscataway NJ USA, pp. 309-316, IEEE Winter Conference on Applications of Computer Vision 2014, Steamboat Springs, United States of America, 24/03/14. https://doi.org/10.1109/WACV.2014.6836084

Local inter-session variability modelling for object classification. / Anantharajah, Kaneswaran; Ge, Zong Yuan; McCool, Chris; Denman, Simon; Fookes, Clinton; Corke, Peter; Tjondronegoro, Dian; Sridharan, Sridha.

2014 IEEE Winter Conference on Applications of Computer Vision, WACV 2014. ed. / Walter Scheirer; Ruigang Yang; Charles Stewart. Piscataway NJ USA : IEEE, Institute of Electrical and Electronics Engineers, 2014. p. 309-316 6836084.

Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

TY - GEN

T1 - Local inter-session variability modelling for object classification

AU - Anantharajah, Kaneswaran

AU - Ge, Zong Yuan

AU - McCool, Chris

AU - Denman, Simon

AU - Fookes, Clinton

AU - Corke, Peter

AU - Tjondronegoro, Dian

AU - Sridharan, Sridha

PY - 2014/1/1

Y1 - 2014/1/1

N2 - Object classification is plagued by the issue of session variation. Session variation describes any variation that makes one instance of an object look different to another, for instance due to pose or illumination variation. Recent work in the challenging task of face verification has shown that session variability modelling provides a mechanism to overcome some of these limitations. However, for computer vision purposes, it has only been applied in the limited setting of face verification. In this paper we propose a local region based intersession variability (ISV) modelling approach, and apply it to challenging real-world data. We propose a region based session variability modelling approach so that local session variations can be modelled, termed Local ISV. We then demonstrate the efficacy of this technique on a challenging real-world fish image database which includes images taken underwater, providing significant real-world session variations. This Local ISV approach provides a relative performance improvement of, on average, 23% on the challenging MOBIO, Multi-PIE and SCface face databases. It also provides a relative performance improvement of 35% on our challenging fish image dataset.

AB - Object classification is plagued by the issue of session variation. Session variation describes any variation that makes one instance of an object look different to another, for instance due to pose or illumination variation. Recent work in the challenging task of face verification has shown that session variability modelling provides a mechanism to overcome some of these limitations. However, for computer vision purposes, it has only been applied in the limited setting of face verification. In this paper we propose a local region based intersession variability (ISV) modelling approach, and apply it to challenging real-world data. We propose a region based session variability modelling approach so that local session variations can be modelled, termed Local ISV. We then demonstrate the efficacy of this technique on a challenging real-world fish image database which includes images taken underwater, providing significant real-world session variations. This Local ISV approach provides a relative performance improvement of, on average, 23% on the challenging MOBIO, Multi-PIE and SCface face databases. It also provides a relative performance improvement of 35% on our challenging fish image dataset.

UR - http://www.scopus.com/inward/record.url?scp=84904704882&partnerID=8YFLogxK

U2 - 10.1109/WACV.2014.6836084

DO - 10.1109/WACV.2014.6836084

M3 - Conference Paper

SN - 9781479949847

SP - 309

EP - 316

BT - 2014 IEEE Winter Conference on Applications of Computer Vision, WACV 2014

A2 - Scheirer, Walter

A2 - Yang, Ruigang

A2 - Stewart, Charles

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - Piscataway NJ USA

ER -

Anantharajah K, Ge ZY, McCool C, Denman S, Fookes C, Corke P et al. Local inter-session variability modelling for object classification. In Scheirer W, Yang R, Stewart C, editors, 2014 IEEE Winter Conference on Applications of Computer Vision, WACV 2014. Piscataway NJ USA: IEEE, Institute of Electrical and Electronics Engineers. 2014. p. 309-316. 6836084 https://doi.org/10.1109/WACV.2014.6836084