Person-similarity weighted feature for expression recognition

Huachun Tan; Yu Jin Zhang

doi:10.1007/978-3-540-76390-1_70

Person-similarity weighted feature for expression recognition

Huachun Tan^*, Yu Jin Zhang

^*Corresponding author for this work

Advanced Research Institute of Multidisciplinary Science

Tsinghua University

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

1 Citation (Scopus)

Abstract

In this paper, a new method to extract person-independent expression feature based on HOSVD (Higher-Order Singular Value Decomposition) is proposed for facial expression recognition. With the assumption that similar persons have similar facial expression appearance and shape, person-similarity weighted expression feature is used to estimate the expression feature of the test person. As a result, the estimated expression feature can reduce the influence of individual caused by insufficient training data and becomes less person-dependent, and can be more robust to new persons. The proposed method has been tested on Cohn-Kanade facial expression database and Japanese Female Facial Expression (JAFFE) database. Person-independent experimental results show the efficiency of the proposed method.

Original language	English
Title of host publication	Computer Vision - ACCV 2007 - 8th Asian Conference on Computer Vision, Proceedings
Publisher	Springer Verlag
Pages	712-721
Number of pages	10
Edition	PART 2
ISBN (Print)	9783540763895
DOIs	https://doi.org/10.1007/978-3-540-76390-1_70
Publication status	Published - 2007
Event	8th Asian Conference on Computer Vision, ACCV 2007 - Tokyo, Japan Duration: 18 Nov 2007 → 22 Nov 2007

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Number	PART 2
Volume	4844 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	8th Asian Conference on Computer Vision, ACCV 2007
Country/Territory	Japan
City	Tokyo
Period	18/11/07 → 22/11/07

Access to Document

10.1007/978-3-540-76390-1_70

Cite this

Tan, H., & Zhang, Y. J. (2007). Person-similarity weighted feature for expression recognition. In Computer Vision - ACCV 2007 - 8th Asian Conference on Computer Vision, Proceedings (PART 2 ed., pp. 712-721). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 4844 LNCS, No. PART 2). Springer Verlag. https://doi.org/10.1007/978-3-540-76390-1_70

@inproceedings{709cf4045e264fe682df5ac9bad7eefb,

title = "Person-similarity weighted feature for expression recognition",

abstract = "In this paper, a new method to extract person-independent expression feature based on HOSVD (Higher-Order Singular Value Decomposition) is proposed for facial expression recognition. With the assumption that similar persons have similar facial expression appearance and shape, person-similarity weighted expression feature is used to estimate the expression feature of the test person. As a result, the estimated expression feature can reduce the influence of individual caused by insufficient training data and becomes less person-dependent, and can be more robust to new persons. The proposed method has been tested on Cohn-Kanade facial expression database and Japanese Female Facial Expression (JAFFE) database. Person-independent experimental results show the efficiency of the proposed method.",

author = "Huachun Tan and Zhang, {Yu Jin}",

year = "2007",

doi = "10.1007/978-3-540-76390-1_70",

language = "English",

isbn = "9783540763895",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

number = "PART 2",

pages = "712--721",

booktitle = "Computer Vision - ACCV 2007 - 8th Asian Conference on Computer Vision, Proceedings",

address = "Germany",

edition = "PART 2",

note = "8th Asian Conference on Computer Vision, ACCV 2007 ; Conference date: 18-11-2007 Through 22-11-2007",

}

Tan, H & Zhang, YJ 2007, Person-similarity weighted feature for expression recognition. in Computer Vision - ACCV 2007 - 8th Asian Conference on Computer Vision, Proceedings. PART 2 edn, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), no. PART 2, vol. 4844 LNCS, Springer Verlag, pp. 712-721, 8th Asian Conference on Computer Vision, ACCV 2007, Tokyo, Japan, 18/11/07. https://doi.org/10.1007/978-3-540-76390-1_70

Person-similarity weighted feature for expression recognition. / Tan, Huachun; Zhang, Yu Jin.
Computer Vision - ACCV 2007 - 8th Asian Conference on Computer Vision, Proceedings. PART 2. ed. Springer Verlag, 2007. p. 712-721 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 4844 LNCS, No. PART 2).

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

TY - GEN

T1 - Person-similarity weighted feature for expression recognition

AU - Tan, Huachun

AU - Zhang, Yu Jin

PY - 2007

Y1 - 2007

N2 - In this paper, a new method to extract person-independent expression feature based on HOSVD (Higher-Order Singular Value Decomposition) is proposed for facial expression recognition. With the assumption that similar persons have similar facial expression appearance and shape, person-similarity weighted expression feature is used to estimate the expression feature of the test person. As a result, the estimated expression feature can reduce the influence of individual caused by insufficient training data and becomes less person-dependent, and can be more robust to new persons. The proposed method has been tested on Cohn-Kanade facial expression database and Japanese Female Facial Expression (JAFFE) database. Person-independent experimental results show the efficiency of the proposed method.

AB - In this paper, a new method to extract person-independent expression feature based on HOSVD (Higher-Order Singular Value Decomposition) is proposed for facial expression recognition. With the assumption that similar persons have similar facial expression appearance and shape, person-similarity weighted expression feature is used to estimate the expression feature of the test person. As a result, the estimated expression feature can reduce the influence of individual caused by insufficient training data and becomes less person-dependent, and can be more robust to new persons. The proposed method has been tested on Cohn-Kanade facial expression database and Japanese Female Facial Expression (JAFFE) database. Person-independent experimental results show the efficiency of the proposed method.

UR - http://www.scopus.com/inward/record.url?scp=38349066304&partnerID=8YFLogxK

U2 - 10.1007/978-3-540-76390-1_70

DO - 10.1007/978-3-540-76390-1_70

M3 - Conference contribution

AN - SCOPUS:38349066304

SN - 9783540763895

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 712

EP - 721

BT - Computer Vision - ACCV 2007 - 8th Asian Conference on Computer Vision, Proceedings

PB - Springer Verlag

T2 - 8th Asian Conference on Computer Vision, ACCV 2007

Y2 - 18 November 2007 through 22 November 2007

ER -

Tan H, Zhang YJ. Person-similarity weighted feature for expression recognition. In Computer Vision - ACCV 2007 - 8th Asian Conference on Computer Vision, Proceedings. PART 2 ed. Springer Verlag. 2007. p. 712-721. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 2). doi: 10.1007/978-3-540-76390-1_70

Person-similarity weighted feature for expression recognition

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this