# American Institute of Mathematical Sciences

February  2019, 2(1): 43-53. doi: 10.3934/mfc.2019004

## Eliminating other-race effect for multi-ethnic facial expression recognition

 1 Dalian Key Lab of Digital Technology for National Culture, College of Computer Science and Engineering, Dalian Minzu University, Dalian 116600, Liaoning, China 2 Department of Computing, Curtin University, Kent Street, Perth, WA 6102, Australia

* Corresponding author: Xiaodong Duan

Published  March 2019

Fund Project: This work is supported by National Natural Science Foundation of China (Grant No.61672132), Science and Technology Foundation of Liaoning Province of China (Grant No.20170520234)ìNational Natural Science Foundation of China (Grant No.61602321), and Natural Science Fund Project of Liaoning Province (No.20170540694).

It has been noticed that the performance of multi-ethnic facial expression recognition is affected by other-race effect significantly. Though this phenomenon has been noticed by psychologists and computer vision researchers for decades, the mechanism of other-race effect is still unknown and few work has been done to compensate or remove this effect. This work proposes an ICA-based method to eliminate the other-race effect in automatic 3D facial expression recognition. Firstly, the depth features are extracted from 3D local facial patches, and independent component analysis is applied to project the features into a subspace in which the projected features are mutually independent. The ethnic-related features and expression-related features are supposed to be separated in ICA subspace. Hence, ethnic-sensitive features are then determined by an entropy-based feature selection method and discarded to depress their influence on facial expression recognition. The proposed method is evaluated on benchmark BU-3DFE database, and the experimental results reveal that the influence caused by other-race effect can be suppressed effectively with the proposed method.

Citation: Mingliang Xue, Xiaodong Duan, Wanquan Liu. Eliminating other-race effect for multi-ethnic facial expression recognition. Mathematical Foundations of Computing, 2019, 2 (1) : 43-53. doi: 10.3934/mfc.2019004
##### References:

show all references

##### References:
The average performance of facial expression recognition on East-Asian individuals when the ethnic-related features are removed gradually
The confusion matrix of multi-ethnic facial expression recognition before(a) and after(b) ethnic-related feature elimination based on East-Asian individuals
The average performance of facial expression recognition on White individuals when the ethnic-related features are removed gradually
The confusion matrix of multi-ethnic facial expression recognition before(a) and after(b) ethnic-related feature elimination based on White individuals
The ethnicity distribution of BU-3DFE database
 Ethnicity Sample Size Number of 3D Faces White 51 1224 East-Asian 24 576 Black 9 216 Hispanic-Latino 8 192 Indian 6 144 Middle-East Asian 2 48
 Ethnicity Sample Size Number of 3D Faces White 51 1224 East-Asian 24 576 Black 9 216 Hispanic-Latino 8 192 Indian 6 144 Middle-East Asian 2 48
 [1] Jian Zhao, Fang Deng, Jian Jia, Chunmeng Wu, Haibo Li, Yuan Shi, Shunli Zhang. A new face feature point matrix based on geometric features and illumination models for facial attraction analysis. Discrete & Continuous Dynamical Systems - S, 2019, 12 (4&5) : 1065-1072. doi: 10.3934/dcdss.2019073 [2] Qiang Yin, Gongfa Li, Jianguo Zhu. Research on the method of step feature extraction for EOD robot based on 2D laser radar. Discrete & Continuous Dynamical Systems - S, 2015, 8 (6) : 1415-1421. doi: 10.3934/dcdss.2015.8.1415 [3] Jianguo Dai, Wenxue Huang, Yuanyi Pan. A category-based probabilistic approach to feature selection. Big Data & Information Analytics, 2017, 2 (5) : 1-8. doi: 10.3934/bdia.2017020 [4] Zhi Guo Feng, K. F. Cedric Yiu, K.L. Mak. Feature extraction of the patterned textile with deformations via optimal control theory. Discrete & Continuous Dynamical Systems - B, 2011, 16 (4) : 1055-1069. doi: 10.3934/dcdsb.2011.16.1055 [5] Yunmei Lu, Mingyuan Yan, Meng Han, Qingliang Yang, Yanqing Zhang. Privacy preserving feature selection and Multiclass Classification for horizontally distributed data. Mathematical Foundations of Computing, 2018, 1 (4) : 331-348. doi: 10.3934/mfc.2018016 [6] Mohamed A. Tawhid, Kevin B. Dsouza. Hybrid binary dragonfly enhanced particle swarm optimization algorithm for solving feature selection problems. Mathematical Foundations of Computing, 2018, 1 (2) : 181-200. doi: 10.3934/mfc.2018009 [7] Mohammed Abdulrazaq Kahya, Suhaib Abduljabbar Altamir, Zakariya Yahya Algamal. Improving whale optimization algorithm for feature selection with a time-varying transfer function. Numerical Algebra, Control & Optimization, 2019, 0 (0) : 0-0. doi: 10.3934/naco.2020017 [8] Makram Hamouda, Chang-Yeol Jung, Roger Temam. Asymptotic analysis for the 3D primitive equations in a channel. Discrete & Continuous Dynamical Systems - S, 2013, 6 (2) : 401-422. doi: 10.3934/dcdss.2013.6.401 [9] Yong Zhou. Remarks on regularities for the 3D MHD equations. Discrete & Continuous Dynamical Systems - A, 2005, 12 (5) : 881-886. doi: 10.3934/dcds.2005.12.881 [10] Hyeong-Ohk Bae, Bum Ja Jin. Estimates of the wake for the 3D Oseen equations. Discrete & Continuous Dynamical Systems - B, 2008, 10 (1) : 1-18. doi: 10.3934/dcdsb.2008.10.1 [11] Indranil SenGupta, Weisheng Jiang, Bo Sun, Maria Christina Mariani. Superradiance problem in a 3D annular domain. Conference Publications, 2011, 2011 (Special) : 1309-1318. doi: 10.3934/proc.2011.2011.1309 [12] Giovanny Guerrero, José Antonio Langa, Antonio Suárez. Biodiversity and vulnerability in a 3D mutualistic system. Discrete & Continuous Dynamical Systems - A, 2014, 34 (10) : 4107-4126. doi: 10.3934/dcds.2014.34.4107 [13] Sadek Gala. A new regularity criterion for the 3D MHD equations in $R^3$. Communications on Pure & Applied Analysis, 2012, 11 (3) : 973-980. doi: 10.3934/cpaa.2012.11.973 [14] Marko Filipović, Ivica Kopriva. A comparison of dictionary based approaches to inpainting and denoising with an emphasis to independent component analysis learned dictionaries. Inverse Problems & Imaging, 2011, 5 (4) : 815-841. doi: 10.3934/ipi.2011.5.815 [15] Danuta Gaweł, Krzysztof Fujarewicz. On the sensitivity of feature ranked lists for large-scale biological data. Mathematical Biosciences & Engineering, 2013, 10 (3) : 667-690. doi: 10.3934/mbe.2013.10.667 [16] Lok Ming Lui, Yalin Wang, Tony F. Chan, Paul M. Thompson. Brain anatomical feature detection by solving partial differential equations on general manifolds. Discrete & Continuous Dynamical Systems - B, 2007, 7 (3) : 605-618. doi: 10.3934/dcdsb.2007.7.605 [17] Jiahong Wu. Regularity results for weak solutions of the 3D MHD equations. Discrete & Continuous Dynamical Systems - A, 2004, 10 (1&2) : 543-556. doi: 10.3934/dcds.2004.10.543 [18] Gabriel Deugoue. Approximation of the trajectory attractor of the 3D MHD System. Communications on Pure & Applied Analysis, 2013, 12 (5) : 2119-2144. doi: 10.3934/cpaa.2013.12.2119 [19] Alp Eden, Varga K. Kalantarov. 3D convective Cahn--Hilliard equation. Communications on Pure & Applied Analysis, 2007, 6 (4) : 1075-1086. doi: 10.3934/cpaa.2007.6.1075 [20] Chongsheng Cao. Sufficient conditions for the regularity to the 3D Navier-Stokes equations. Discrete & Continuous Dynamical Systems - A, 2010, 26 (4) : 1141-1151. doi: 10.3934/dcds.2010.26.1141

Impact Factor:

## Tools

Article outline

Figures and Tables