Projective complex matrix factorization for facial expression recognition

被引:0
|
作者
Viet-Hang Duong
Yuan-Shan Lee
Jian-Jiun Ding
Bach-Tung Pham
Manh-Quan Bui
Pham The Bao
Jia-Ching Wang
机构
[1] National Central University,Department of Computer Science and Information Engineering
[2] National Taiwan University,Graduate Institute of Communication Engineering
[3] University of Science,Faculty of Mathematics and Computer Science
来源
EURASIP Journal on Advances in Signal Processing | / 2018卷
关键词
Complex matrix factorization; Facial expression recognition; Nonnegative matrix factorization; Projected gradient descent;
D O I
暂无
中图分类号
学科分类号
摘要
In this paper, a dimensionality reduction method applied on facial expression recognition is investigated. An unsupervised learning framework, projective complex matrix factorization (proCMF), is introduced to project high-dimensional input facial images into a lower dimension subspace. The proCMF model is related to both the conventional projective nonnegative matrix factorization (proNMF) and the cosine dissimilarity metric in the simple manner by transforming real data into the complex domain. A projective matrix is then found through solving an unconstraint complex optimization problem. The gradient descent method was utilized to optimize a complex cost function. Extensive experiments carried on the extended Cohn-Kanade and the JAFFE databases show that the proposed proCMF model provides even better performance than state-of-the-art methods for facial expression recognition.
引用
收藏
相关论文
共 50 条
  • [11] A Novel Discriminant Non-Negative Matrix Factorization and its Application to Facial Expression Recognition
    Zhu, Yan-Li
    Chen, Jun
    Qu, Pei-xin
    SMART MATERIALS AND INTELLIGENT SYSTEMS, PTS 1 AND 2, 2011, 143-144 : 129 - 133
  • [12] Appearance factorization based facial expression recognition and synthesis
    Abboud, B
    Davoine, F
    PROCEEDINGS OF THE 17TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION, VOL 4, 2004, : 163 - 166
  • [13] FACIAL EXPRESSION RECOGNITION BASED ON GRAPH-PRESERVING SPARSE NON-NEGATIVE MATRIX FACTORIZATION
    Zhi, Ruicong
    Flierl, Markus
    Ruan, Qiuqi
    Kleijn, Bastiaan
    2009 16TH IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, VOLS 1-6, 2009, : 3293 - +
  • [14] Transferable discriminative non-negative matrix factorization for cross-database facial expression recognition
    Ji, Liang
    Song, Peng
    Zhang, Wenjing
    DIGITAL SIGNAL PROCESSING, 2022, 123
  • [15] HUMAN FACIAL EXPRESSION RECOGNITION USING TWO-DIMENSIONAL NON-NEGATIVE MATRIX FACTORIZATION
    Salim, William
    Santika, Diaz D.
    INTERNATIONAL CONFERENCE ON ADVANCES SCIENCE AND CONTEMPORARY ENGINEERING 2012, 2012, 50 : 758 - 767
  • [16] Graph regularized projective non-negative matrix factorization for face recognition
    Yu, Z. (yuzz@jlu.edu.cn), 2013, Binary Information Press, P.O. Box 162, Bethel, CT 06801-0162, United States (09):
  • [17] Facial Expression Recognition based on Orthogonal Nonnegative CP Factorization
    Liu, Chang
    He, Kun
    Zhou, Ji-Liu
    Zhu, Yan-Li
    SMART MATERIALS AND INTELLIGENT SYSTEMS, PTS 1 AND 2, 2011, 143-144 : 111 - +
  • [18] A PROJECTIVE APPROACH TO NONNEGATIVE MATRIX FACTORIZATION
    Groetzner, Patrick
    ELECTRONIC JOURNAL OF LINEAR ALGEBRA, 2021, 37 : 583 - 597
  • [19] Projective Nonnegative Matrix Factorization with α-Divergence
    Yang, Zhirong
    Oja, Erkki
    ARTIFICIAL NEURAL NETWORKS - ICANN 2009, PT I, 2009, 5768 : 20 - 29
  • [20] Generalized Discriminant Orthogonal Nonnegative Tensor Factorization for Facial Expression Recognition
    Zhang XiuJun
    Liu Chang
    SCIENTIFIC WORLD JOURNAL, 2014,