Enhancing ELM-based facial image classification by exploiting multiple facial views

Alexandros Iosifidis; Anastasios Tefas; Ioannis Pitas

doi:10.1016/j.procs.2015.05.440

Enhancing ELM-based facial image classification by exploiting multiple facial views

Alexandros Iosifidis
, Anastasios Tefas
, Ioannis Pitas

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

6 Citations (Scopus)

Abstract

In this paper, we investigate the effectiveness of the Extreme Learning Machine (ELM) network in facial image classification. In order to enhance performance, we exploit knowledge related to the human face structure. We train a multi-view ELM network by employing automatically created facial regions of interest to this end. By jointly learning the network parameters and optimized network output combination weights, each facial region appropriately contributes to the final classification result. Experimental results on three publicly available databases show that the proposed approach outperforms facial image classification based on a single facial representation and on other facial region combination schemes.

Original language	English
Title of host publication	Procedia Computer Science
Publisher	Elsevier
Pages	2814-2821
Number of pages	8
Volume	51
DOIs	https://doi.org/10.1016/j.procs.2015.05.440
Publication status	Published - 2015
Publication type	A4 Article in conference proceedings
Event	15th Annual International Conference on Computational Science, ICCS 2015 - Reykjavik, Iceland Duration: 1 Jun 2015 → 3 Jun 2015

Conference

Conference	15th Annual International Conference on Computational Science, ICCS 2015
Country/Territory	Iceland
City	Reykjavik
Period	1/06/15 → 3/06/15

Keywords

Extreme Learning Machine
Facial image classification
Multi-view learning

ASJC Scopus subject areas

General Computer Science

Access to Document

10.1016/j.procs.2015.05.440

Cite this

@inproceedings{af32f44fc6cc424ba41e360cf4d8fd88,

title = "Enhancing ELM-based facial image classification by exploiting multiple facial views",

abstract = "In this paper, we investigate the effectiveness of the Extreme Learning Machine (ELM) network in facial image classification. In order to enhance performance, we exploit knowledge related to the human face structure. We train a multi-view ELM network by employing automatically created facial regions of interest to this end. By jointly learning the network parameters and optimized network output combination weights, each facial region appropriately contributes to the final classification result. Experimental results on three publicly available databases show that the proposed approach outperforms facial image classification based on a single facial representation and on other facial region combination schemes.",

keywords = "Extreme Learning Machine, Facial image classification, Multi-view learning",

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

year = "2015",

doi = "10.1016/j.procs.2015.05.440",

language = "English",

volume = "51",

pages = "2814--2821",

booktitle = "Procedia Computer Science",

publisher = "Elsevier",

address = "Netherlands",

note = "15th Annual International Conference on Computational Science, ICCS 2015 ; Conference date: 01-06-2015 Through 03-06-2015",

}

TY - GEN

T1 - Enhancing ELM-based facial image classification by exploiting multiple facial views

AU - Iosifidis, Alexandros

AU - Tefas, Anastasios

AU - Pitas, Ioannis

PY - 2015

Y1 - 2015

N2 - In this paper, we investigate the effectiveness of the Extreme Learning Machine (ELM) network in facial image classification. In order to enhance performance, we exploit knowledge related to the human face structure. We train a multi-view ELM network by employing automatically created facial regions of interest to this end. By jointly learning the network parameters and optimized network output combination weights, each facial region appropriately contributes to the final classification result. Experimental results on three publicly available databases show that the proposed approach outperforms facial image classification based on a single facial representation and on other facial region combination schemes.

AB - In this paper, we investigate the effectiveness of the Extreme Learning Machine (ELM) network in facial image classification. In order to enhance performance, we exploit knowledge related to the human face structure. We train a multi-view ELM network by employing automatically created facial regions of interest to this end. By jointly learning the network parameters and optimized network output combination weights, each facial region appropriately contributes to the final classification result. Experimental results on three publicly available databases show that the proposed approach outperforms facial image classification based on a single facial representation and on other facial region combination schemes.

KW - Extreme Learning Machine

KW - Facial image classification

KW - Multi-view learning

U2 - 10.1016/j.procs.2015.05.440

DO - 10.1016/j.procs.2015.05.440

M3 - Conference contribution

AN - SCOPUS:84939545786

VL - 51

SP - 2814

EP - 2821

BT - Procedia Computer Science

PB - Elsevier

T2 - 15th Annual International Conference on Computational Science, ICCS 2015

Y2 - 1 June 2015 through 3 June 2015

ER -

Enhancing ELM-based facial image classification by exploiting multiple facial views

Abstract

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this