Semi-supervised learning for facial expression recognition

Ira Cohen; Nicu Sebe; Fabio G. Cozman; Thomas S. Huang

doi:10.1145/973264.973268

Semi-supervised learning for facial expression recognition

Ira Cohen, Nicu Sebe, Fabio G. Cozman, Thomas S. Huang

Research output: Contribution to conference › Conference Paper

22 Scopus citations

Abstract

Copyright 2003 ACM. Automatic classification by machines is one of the basic tasks required in any pattern recognition and human computer interaction applications. In this paper, we discuss training probabilistic classifiers with labeled and unlabeled data. We provide an analysis which shows under what conditions unlabeled data can be used in learning to improve classification performance. We discuss the implications of this analysis to a specific type of probabilistic classifiers, Bayesian networks, and propose a structure learning algorithm that can utilize unlabeled data to improve classification. Finally, we show how the resulting algorithms are successfully employed in a facial expression recognition application.

Original language	American English
Pages	17-22
Number of pages	6
DOIs	https://doi.org/10.1145/973264.973268
State	Published - 7 Nov 2003
Externally published	Yes
Event	Proceedings of the 5th ACM SIGMM International Workshop on Multimedia Information Retrieval, MIR 2003 - Duration: 7 Nov 2003 → …

Conference

Conference	Proceedings of the 5th ACM SIGMM International Workshop on Multimedia Information Retrieval, MIR 2003
Period	7/11/03 → …

Access to Document

10.1145/973264.973268

Cite this

@conference{bf81529cb85e405a85bafa2a6d78963e,

title = "Semi-supervised learning for facial expression recognition",

abstract = "Copyright 2003 ACM. Automatic classification by machines is one of the basic tasks required in any pattern recognition and human computer interaction applications. In this paper, we discuss training probabilistic classifiers with labeled and unlabeled data. We provide an analysis which shows under what conditions unlabeled data can be used in learning to improve classification performance. We discuss the implications of this analysis to a specific type of probabilistic classifiers, Bayesian networks, and propose a structure learning algorithm that can utilize unlabeled data to improve classification. Finally, we show how the resulting algorithms are successfully employed in a facial expression recognition application.",

author = "Ira Cohen and Nicu Sebe and Cozman, {Fabio G.} and Huang, {Thomas S.}",

year = "2003",

month = nov,

day = "7",

doi = "10.1145/973264.973268",

language = "American English",

pages = "17--22",

note = "Proceedings of the 5th ACM SIGMM International Workshop on Multimedia Information Retrieval, MIR 2003 ; Conference date: 07-11-2003",

}

TY - CONF

T1 - Semi-supervised learning for facial expression recognition

AU - Cohen, Ira

AU - Sebe, Nicu

AU - Cozman, Fabio G.

AU - Huang, Thomas S.

PY - 2003/11/7

Y1 - 2003/11/7

N2 - Copyright 2003 ACM. Automatic classification by machines is one of the basic tasks required in any pattern recognition and human computer interaction applications. In this paper, we discuss training probabilistic classifiers with labeled and unlabeled data. We provide an analysis which shows under what conditions unlabeled data can be used in learning to improve classification performance. We discuss the implications of this analysis to a specific type of probabilistic classifiers, Bayesian networks, and propose a structure learning algorithm that can utilize unlabeled data to improve classification. Finally, we show how the resulting algorithms are successfully employed in a facial expression recognition application.

AB - Copyright 2003 ACM. Automatic classification by machines is one of the basic tasks required in any pattern recognition and human computer interaction applications. In this paper, we discuss training probabilistic classifiers with labeled and unlabeled data. We provide an analysis which shows under what conditions unlabeled data can be used in learning to improve classification performance. We discuss the implications of this analysis to a specific type of probabilistic classifiers, Bayesian networks, and propose a structure learning algorithm that can utilize unlabeled data to improve classification. Finally, we show how the resulting algorithms are successfully employed in a facial expression recognition application.

UR - https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84973360336&origin=inward

UR - https://www.scopus.com/inward/citedby.uri?partnerID=HzOxMe3b&scp=84973360336&origin=inward

U2 - 10.1145/973264.973268

DO - 10.1145/973264.973268

M3 - Conference Paper

SP - 17

EP - 22

T2 - Proceedings of the 5th ACM SIGMM International Workshop on Multimedia Information Retrieval, MIR 2003

Y2 - 7 November 2003

ER -

Semi-supervised learning for facial expression recognition

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Cite this