Class-specific nonlinear subspace learning based on optimized class representation

Alexandros Iosifidis; Anastasios Tefas; Ioannis Pitas

doi:10.1109/EUSIPCO.2015.7362833

Class-specific nonlinear subspace learning based on optimized class representation

Alexandros Iosifidis, Anastasios Tefas, Ioannis Pitas

Department of Electrical & Electronic Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

225 Downloads (Pure)

Abstract

In this paper, a new nonlinear subspace learning technique for class-specific data representation based on an optimized class representation is described. An iterative optimization scheme is formulated where both the optimal nonlinear data
projection and the optimal class representation are determined at each optimization step. This approach is tested on human face and action recognition problems, where its performance is compared with that of the standard class-specific subspace learning approach, as well as other nonlinear discriminant subspace learning techniques. Experimental results denote the effectiveness of this new approach, since it consistently outperforms the standard one and outperforms other nonlinear discriminant subspace learning techniques in most cases.

Original language	English
Title of host publication	2015 23rd European Signal Processing Conference (EUSIPCO)
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
Pages	2491-2495
Number of pages	5
ISBN (Electronic)	9780992862633
ISBN (Print)	9781479988518
DOIs	https://doi.org/10.1109/EUSIPCO.2015.7362833
Publication status	Published - 28 Dec 2015
Event	23rd European Signal Processing Conference, EUSIPCO 2015 - Nice, France Duration: 31 Aug 2015 → 4 Sep 2015

Publication series

Name	Proceedings of the European Signal Processing Conference (EUSIPCO)
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
ISSN (Print)	2219-5491

Conference

Conference	23rd European Signal Processing Conference, EUSIPCO 2015
Country/Territory	France
City	Nice
Period	31/08/15 → 4/09/15

Keywords

Class-specific discriminant learning
Nonlinear subspace learning
Action recognition
Face recognition

Access to Document

10.1109/EUSIPCO.2015.7362833

Ioannis Pitas - Class-specific nonlinear subspace learning based on optimized class representation - 2015
This is the author accepted manuscript (AAM). The final published version (version of record) is available online via IEEE at http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=7362833. Please refer to any applicable terms of use of the publisher.
Accepted author manuscript, 76.8 KB

Handle.net

Persistent link

Cite this

Iosifidis, A., Tefas, A., & Pitas, I. (2015). Class-specific nonlinear subspace learning based on optimized class representation. In 2015 23rd European Signal Processing Conference (EUSIPCO) (pp. 2491-2495). (Proceedings of the European Signal Processing Conference (EUSIPCO)). Institute of Electrical and Electronics Engineers (IEEE). https://doi.org/10.1109/EUSIPCO.2015.7362833

@inproceedings{6164347b5c6f4e7c99aeacb6656246c6,

title = "Class-specific nonlinear subspace learning based on optimized class representation",

abstract = "In this paper, a new nonlinear subspace learning technique for class-specific data representation based on an optimized class representation is described. An iterative optimization scheme is formulated where both the optimal nonlinear dataprojection and the optimal class representation are determined at each optimization step. This approach is tested on human face and action recognition problems, where its performance is compared with that of the standard class-specific subspace learning approach, as well as other nonlinear discriminant subspace learning techniques. Experimental results denote the effectiveness of this new approach, since it consistently outperforms the standard one and outperforms other nonlinear discriminant subspace learning techniques in most cases.",

keywords = "Class-specific discriminant learning, Nonlinear subspace learning, Action recognition, Face recognition",

author = "Alexandros Iosifidis and Anastasios Tefas and Ioannis Pitas",

year = "2015",

month = dec,

day = "28",

doi = "10.1109/EUSIPCO.2015.7362833",

language = "English",

isbn = "9781479988518",

series = "Proceedings of the European Signal Processing Conference (EUSIPCO)",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

pages = "2491--2495",

booktitle = "2015 23rd European Signal Processing Conference (EUSIPCO)",

address = "United States",

note = "23rd European Signal Processing Conference, EUSIPCO 2015 ; Conference date: 31-08-2015 Through 04-09-2015",

}

Iosifidis, A, Tefas, A & Pitas, I 2015, Class-specific nonlinear subspace learning based on optimized class representation. in 2015 23rd European Signal Processing Conference (EUSIPCO). Proceedings of the European Signal Processing Conference (EUSIPCO), Institute of Electrical and Electronics Engineers (IEEE), pp. 2491-2495, 23rd European Signal Processing Conference, EUSIPCO 2015, Nice, France, 31/08/15. https://doi.org/10.1109/EUSIPCO.2015.7362833

Class-specific nonlinear subspace learning based on optimized class representation. / Iosifidis, Alexandros; Tefas, Anastasios; Pitas, Ioannis.

2015 23rd European Signal Processing Conference (EUSIPCO). Institute of Electrical and Electronics Engineers (IEEE), 2015. p. 2491-2495 (Proceedings of the European Signal Processing Conference (EUSIPCO)).

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

TY - GEN

T1 - Class-specific nonlinear subspace learning based on optimized class representation

AU - Iosifidis, Alexandros

AU - Tefas, Anastasios

AU - Pitas, Ioannis

PY - 2015/12/28

Y1 - 2015/12/28

N2 - In this paper, a new nonlinear subspace learning technique for class-specific data representation based on an optimized class representation is described. An iterative optimization scheme is formulated where both the optimal nonlinear dataprojection and the optimal class representation are determined at each optimization step. This approach is tested on human face and action recognition problems, where its performance is compared with that of the standard class-specific subspace learning approach, as well as other nonlinear discriminant subspace learning techniques. Experimental results denote the effectiveness of this new approach, since it consistently outperforms the standard one and outperforms other nonlinear discriminant subspace learning techniques in most cases.

AB - In this paper, a new nonlinear subspace learning technique for class-specific data representation based on an optimized class representation is described. An iterative optimization scheme is formulated where both the optimal nonlinear dataprojection and the optimal class representation are determined at each optimization step. This approach is tested on human face and action recognition problems, where its performance is compared with that of the standard class-specific subspace learning approach, as well as other nonlinear discriminant subspace learning techniques. Experimental results denote the effectiveness of this new approach, since it consistently outperforms the standard one and outperforms other nonlinear discriminant subspace learning techniques in most cases.

KW - Class-specific discriminant learning

KW - Nonlinear subspace learning

KW - Action recognition

KW - Face recognition

U2 - 10.1109/EUSIPCO.2015.7362833

DO - 10.1109/EUSIPCO.2015.7362833

M3 - Conference Contribution (Conference Proceeding)

SN - 9781479988518

T3 - Proceedings of the European Signal Processing Conference (EUSIPCO)

SP - 2491

EP - 2495

BT - 2015 23rd European Signal Processing Conference (EUSIPCO)

PB - Institute of Electrical and Electronics Engineers (IEEE)

T2 - 23rd European Signal Processing Conference, EUSIPCO 2015

Y2 - 31 August 2015 through 4 September 2015

ER -

Class-specific nonlinear subspace learning based on optimized class representation

Abstract

Publication series

Conference

Keywords

Access to Document

Handle.net

Fingerprint

Cite this