A new constrained nonnegative matrix factorization for facial expression recognition

Viet Hang Duong; Manh Quan Bui; Pham The Bao; Jia Ching Wang

doi:10.1109/ICOT.2017.8336093

A new constrained nonnegative matrix factorization for facial expression recognition

Viet Hang Duong
, Manh Quan Bui
, Pham The Bao
, Jia Ching Wang

研究成果: 書貢獻/報告類型 › 會議論文篇章 › 同行評審

1 引文斯高帕斯（Scopus）

摘要

A new NMF model, spatial constrained graph sparse nonnegative matrix factorization (SGSNMF) is adopted for facial expression recognition. In this model, the extracted features preserve the topological structure of the original images and achieve sparseness from L₂ constraint on coefficient matrix, meanwhile the base satisfy pixel dispersion penalty. The proposed method takes advantage of the project gradient decent and is based on the alternating nonnegative least square framework. Experiments on two facial expression recognition scenarios that involve a whole face and an occluded face reveal that the proposed algorithm outperforms the prevalent NMF methods.

原文	???core.languages.en_GB???
主出版物標題	Proceedings of the 2017 International Conference on Orange Technologies, ICOT 2017
編輯	Minghui Dong, Lei Wang, Yanfeng Lu, Haizhou Li
發行者	Institute of Electrical and Electronics Engineers Inc.
頁面	79-82
頁數	4
ISBN（電子）	9781538632758
DOIs	https://doi.org/10.1109/ICOT.2017.8336093
出版狀態	已出版 - 2 7月 2017
事件	5th International Conference on Orange Technologies, ICOT 2017 - Singapore, Singapore 持續時間: 8 12月 2017 → 10 12月 2017

出版系列

名字	Proceedings of the 2017 International Conference on Orange Technologies, ICOT 2017
卷	2018-January

???event.eventtypes.event.conference???

???event.eventtypes.event.conference???	5th International Conference on Orange Technologies, ICOT 2017
國家/地區	Singapore
城市	Singapore
期間	8/12/17 → 10/12/17

UN SDG

此研究成果有助於以下永續發展目標

SDG 3 良好的健康和福祉

存取文件

10.1109/ICOT.2017.8336093

引用此

Duong, V. H., Bui, M. Q., Bao, P. T., & Wang, J. C. (2017). A new constrained nonnegative matrix factorization for facial expression recognition. 於 M. Dong, L. Wang, Y. Lu, & H. Li (編輯), Proceedings of the 2017 International Conference on Orange Technologies, ICOT 2017 (頁 79-82). (Proceedings of the 2017 International Conference on Orange Technologies, ICOT 2017; 卷 2018-January). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICOT.2017.8336093

Duong, Viet Hang ; Bui, Manh Quan ; Bao, Pham The 等. / A new constrained nonnegative matrix factorization for facial expression recognition. Proceedings of the 2017 International Conference on Orange Technologies, ICOT 2017. 編輯 / Minghui Dong ; Lei Wang ; Yanfeng Lu ; Haizhou Li. Institute of Electrical and Electronics Engineers Inc., 2017. 頁 79-82 (Proceedings of the 2017 International Conference on Orange Technologies, ICOT 2017).

@inproceedings{9f96352b2f7f4e29875da39be1f12c3b,

title = "A new constrained nonnegative matrix factorization for facial expression recognition",

abstract = "A new NMF model, spatial constrained graph sparse nonnegative matrix factorization (SGSNMF) is adopted for facial expression recognition. In this model, the extracted features preserve the topological structure of the original images and achieve sparseness from L2 constraint on coefficient matrix, meanwhile the base satisfy pixel dispersion penalty. The proposed method takes advantage of the project gradient decent and is based on the alternating nonnegative least square framework. Experiments on two facial expression recognition scenarios that involve a whole face and an occluded face reveal that the proposed algorithm outperforms the prevalent NMF methods.",

keywords = "Facial expression recognition, Graph regularization, Nonnegative matrix factorization, Projectedgradient descent",

author = "Duong, \{Viet Hang\} and Bui, \{Manh Quan\} and Bao, \{Pham The\} and Wang, \{Jia Ching\}",

note = "Publisher Copyright: {\textcopyright} 2017 IEEE.; 5th International Conference on Orange Technologies, ICOT 2017 ; Conference date: 08-12-2017 Through 10-12-2017",

year = "2017",

month = jul,

day = "2",

doi = "10.1109/ICOT.2017.8336093",

language = "???core.languages.en\_GB???",

series = "Proceedings of the 2017 International Conference on Orange Technologies, ICOT 2017",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "79--82",

editor = "Minghui Dong and Lei Wang and Yanfeng Lu and Haizhou Li",

booktitle = "Proceedings of the 2017 International Conference on Orange Technologies, ICOT 2017",

}

Duong, VH, Bui, MQ, Bao, PT & Wang, JC 2017, A new constrained nonnegative matrix factorization for facial expression recognition. 於 M Dong, L Wang, Y Lu & H Li (編輯), Proceedings of the 2017 International Conference on Orange Technologies, ICOT 2017. Proceedings of the 2017 International Conference on Orange Technologies, ICOT 2017, 卷 2018-January, Institute of Electrical and Electronics Engineers Inc., 頁 79-82, 5th International Conference on Orange Technologies, ICOT 2017, Singapore, Singapore, 8/12/17. https://doi.org/10.1109/ICOT.2017.8336093

A new constrained nonnegative matrix factorization for facial expression recognition. / Duong, Viet Hang; Bui, Manh Quan; Bao, Pham The 等.
Proceedings of the 2017 International Conference on Orange Technologies, ICOT 2017. 編輯 / Minghui Dong; Lei Wang; Yanfeng Lu; Haizhou Li. Institute of Electrical and Electronics Engineers Inc., 2017. p. 79-82 (Proceedings of the 2017 International Conference on Orange Technologies, ICOT 2017; 卷 2018-January).

研究成果: 書貢獻/報告類型 › 會議論文篇章 › 同行評審

TY - GEN

T1 - A new constrained nonnegative matrix factorization for facial expression recognition

AU - Duong, Viet Hang

AU - Bui, Manh Quan

AU - Bao, Pham The

AU - Wang, Jia Ching

PY - 2017/7/2

Y1 - 2017/7/2

N2 - A new NMF model, spatial constrained graph sparse nonnegative matrix factorization (SGSNMF) is adopted for facial expression recognition. In this model, the extracted features preserve the topological structure of the original images and achieve sparseness from L2 constraint on coefficient matrix, meanwhile the base satisfy pixel dispersion penalty. The proposed method takes advantage of the project gradient decent and is based on the alternating nonnegative least square framework. Experiments on two facial expression recognition scenarios that involve a whole face and an occluded face reveal that the proposed algorithm outperforms the prevalent NMF methods.

AB - A new NMF model, spatial constrained graph sparse nonnegative matrix factorization (SGSNMF) is adopted for facial expression recognition. In this model, the extracted features preserve the topological structure of the original images and achieve sparseness from L2 constraint on coefficient matrix, meanwhile the base satisfy pixel dispersion penalty. The proposed method takes advantage of the project gradient decent and is based on the alternating nonnegative least square framework. Experiments on two facial expression recognition scenarios that involve a whole face and an occluded face reveal that the proposed algorithm outperforms the prevalent NMF methods.

KW - Facial expression recognition

KW - Graph regularization

KW - Nonnegative matrix factorization

KW - Projectedgradient descent

UR - https://www.scopus.com/pages/publications/85048558545

U2 - 10.1109/ICOT.2017.8336093

DO - 10.1109/ICOT.2017.8336093

M3 - 會議論文篇章

AN - SCOPUS:85048558545

T3 - Proceedings of the 2017 International Conference on Orange Technologies, ICOT 2017

SP - 79

EP - 82

BT - Proceedings of the 2017 International Conference on Orange Technologies, ICOT 2017

A2 - Dong, Minghui

A2 - Wang, Lei

A2 - Lu, Yanfeng

A2 - Li, Haizhou

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 5th International Conference on Orange Technologies, ICOT 2017

Y2 - 8 December 2017 through 10 December 2017

ER -

Duong VH, Bui MQ, Bao PT, Wang JC. A new constrained nonnegative matrix factorization for facial expression recognition. 於 Dong M, Wang L, Lu Y, Li H, 編輯, Proceedings of the 2017 International Conference on Orange Technologies, ICOT 2017. Institute of Electrical and Electronics Engineers Inc. 2017. p. 79-82. (Proceedings of the 2017 International Conference on Orange Technologies, ICOT 2017). doi: 10.1109/ICOT.2017.8336093

A new constrained nonnegative matrix factorization for facial expression recognition

摘要

出版系列

???event.eventtypes.event.conference???

UN SDG

存取文件

指紋

引用此