基于特征级和决策级融合的人脸吸引力评价方法

doi:10.11772/j.issn.1001-9081.2018051040

计算机应用 ›› 2018, Vol. 38 ›› Issue (12): 3607-3611.DOI: 10.11772/j.issn.1001-9081.2018051040

• 应用前沿、交叉与综合 • 上一篇下一篇

基于特征级和决策级融合的人脸吸引力评价方法

李金蔓¹, 汪剑鸣^1,2, 金光浩²

1. 天津工业大学电子与信息工程学院, 天津 300387;
2. 天津工业大学计算机科学与技术学院, 天津 300387

收稿日期:2018-05-21 修回日期:2018-08-27 出版日期:2018-12-10 发布日期:2018-12-15
通讯作者: 金光浩
作者简介:李金蔓(1993-),女,吉林长春人,硕士研究生,主要研究方向:人脸图像处理;汪剑鸣(1974-),男,河北隆尧人,教授,博士,CCF会员,主要研究方向:计算机视觉、智能信息处理;金光浩(1979-),男,吉林延吉人,讲师,博士,CCF会员,主要研究方向:大数据分析、计算机视觉。
基金资助:
国家自然科学基金资助项目（61771340，61302127，61403278）；中国博士后科学基金资助项目（2015M570228）；天津市应用基础与前沿技术研究计划项目（15JCYBJC16600）；天津市自然科学基金资助项目（16JCYBJC42300）；天津市高等学校创新团队培养计划项目（TD13-5032）。

Facial attractiveness evaluation method based on fusion of feature-level and decision-level

LI Jinman¹, WANG Jianming^1,2, JIN Guanghao²

1. School of Electronics and Information Engineering, Tianjin Polytechinic University, Tianjin 300387, China;
2. School of Computer Science and Technology, Tianjin Polytechinic University, Tianjin 300387, China

Received:2018-05-21 Revised:2018-08-27 Online:2018-12-10 Published:2018-12-15
Contact: 金光浩
Supported by:
This work is partially supported by the National Natural Science Foundation of China (61771340, 61302127, 61403278), the Postdoctoral Science Foundation of China (2015M570228), the Tianjin Research Program of Application Foundation and Advanced Technology (15JCYBJC16600), the Natural Science Foundation of Tianjin (16JCYBJC42300), the Tianjin Higher Education Innovation Team Training Program (TD13-5032).

摘要/Abstract

摘要： 在个性化的人脸吸引力的研究中，由于特征缺失和对于大众审美的影响因素考虑不足，导致预测个人偏好无法到达很高的预测精度。为了提高预测精度，提出了一个基于特征级和决策级信息融合的个性化人脸吸引力预测框架。首先，将代表不同人脸美丽特征的客观特性融合到一起，利用特征选择算法挑选出具有代表性的人脸吸引力特征，并利用不同的信息融合策略将人脸局部、全局特征融合起来；然后，将传统的人脸特征与通过深度网络自动提取的特征融合起来。同时，提出多种融合策略进行对比，将代表着大众审美偏好的评分信息与代表个人偏好的个性化评分信息进行决策级融合，最终实现个性化的人脸吸引力预测评分。实验结果表明，相比现有针对个性化人脸吸引力评价研究的算法，所提的多层次融合方法在预测精度方面有显著的提升，能够达到Pearson相关系数0.9以上。该方法可用于个性化推荐、人脸美化等领域。

关键词: 人脸吸引力, 特征级融合, 决策级融合, 个性化信息, 共识性信息

Abstract: In the study of personalized facial attractiveness, due to lack of features and insufficient consideration of the influence factors of public aesthetics, the prediction of personal preferences cannot reach high prediction accuracy. In order to improve the prediction accuracy, a new personalized facial attractiveness prediction framework based on feature-level and decision-level information fusion was proposed. Firstly, the objective characteristics of different facial beauty features were fused together, and the representative facial attractive features were selected by a feature selection algorithm, the local and global features of face were fused by different information fusion strategies. Then, the traditional facial features were fused with the features extracted automatically through deep networks. At the same time, a variety of fusion strategies were proposed for comparison. The score information representing the public aesthetic preferences and the personalized score information representing the individual preferences were fused at the decision level. Finally, the personalized facial attractiveness prediction score was obtained. The experimental results show that, compared with the existing algorithms for personalized facial attractiveness evaluation, the proposed multi-level fusion method has a significant improvement in prediction accuracy, and can achieve the Pearson correlation coefficient more than 0.9. The proposed method can be used in the fields of personalized recommendation and face beautification.

Key words: facial attractiveness, feature-level fusion, decision-level fusion, personalized information, consensual information

中图分类号:

TP391

李金蔓, 汪剑鸣, 金光浩. 基于特征级和决策级融合的人脸吸引力评价方法[J]. 计算机应用, 2018, 38(12): 3607-3611.

LI Jinman, WANG Jianming, JIN Guanghao. Facial attractiveness evaluation method based on fusion of feature-level and decision-level[J]. Journal of Computer Applications, 2018, 38(12): 3607-3611.

参考文献

[1] KIM C H, OH J Y, CHOI C H. Combined subspace method using global and local features for face recognition[C]//Proceedings of the 2005 IEEE International Joint Conference on Neural Networks. Piscataway, NJ:IEEE, 2005:2030-2035.
[2] EISENTHAL Y, DROR G, RUPPIN E. Facial attractiveness:beauty and the machine[J]. Neural Computation, 2006, 18(1):119-142.
[3] MAO H Y, JIN L W, DU M H. Automatic classification of Chinese female facial beauty using support vector machine[C]//Proceedings of the 2009 IEEE International Conference on Systems, Man and Cybernetics. Piscataway, NJ:IEEE, 2009:4842-4846.
[4] ZHANG D, CHEN F M, XU Y. Data-driven facial beauty analysis:prediction, retrieval and manipulation[M]//Computer Models for Facial Beauty Analysis. Cham:Springer, 2016:217-234.
[5] GAN J Y, LI L C, ZHAI Y K. Facial beauty prediction model based on self-taught learning and convolutional restricted Boltzmann machine[C]//Proceedings of the 2014 IEEE International Conference on Machine Learning and Cybernetics. Piscataway, NJ:IEEE, 2014:844-849.
[6] GAN J Y, ZHOU L, ZHAI Y K. A study for facial beauty prediction model[C]//Proceedings of the 2015 IEEE International Conference on Wavelet Analysis and Pattern Recognition. Piscataway, NJ:IEEE, 2015:8-13.
[7] SIMONYAN K, ZISSERMAN A. Two-stream convolutional networks for action recognition in videos[EB/OL].[2018-03-26]. http://papers.nips.cc/paper/5353-two-stream-convolutional-networks-for-action-recognition-in-videos.pdf.
[8] 何刚,霍宏,方涛.兼顾特征级和决策级融合的场景分类[J].计算机应用,2016,36(5):1262-1266.(HE G, HUO H, FANG T. Scene classification based ono feature-level and decision-level fusion[J]. Journal of Computer Applications, 2016, 36(5):1262-1266.)
[9] 毛慧芸,金连文,杜明辉.基于几何特征及C4.5的人脸美丽分类方法[J].模式识别与人工智能,2010,23(6):809-814.(MAO H Y, JIN L W, DU M H. Facial beauty classification based on geometric features and C4.5[J]. Pattern Recognition and Artificial Intelligence, 2010, 23(6):809-814.)
[10] LIU C J, WECHSLER H. Gabor feature based classification using the enhanced Fisher linear discriminant model for face recognition[J]. IEEE Transactions on Image Processing, 2002, 11(4):467-476.
[11] BELHUMEUR P N, HESPANHA J P, KRIEGMAN D J. Eigenfaces vs. Fisherfaces:recognition using class specific linear projection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1997, 19(7):711-720.
[12] 唐斌,肖先赐.基于FFT的欠采样带宽信号的频率无模糊估计[J].数据采集与处理,1998,13(1):17-20.(TANG B, XIAO X C. Undersampling wide-band signal frequency unambiguous estimation based on FFT[J]. Journal of Data Acquisition and Processing, 1998, 13(1):17-20.)
[13] XIE D R, LIANG L Y, JIN L W, et al. SCUT-FBP:a benchmark dataset for facial beauty perception[C]//Proceedings of the 2015 IEEE International Conference on Systems, Man, and Cybernetics. Piscataway, NJ:IEEE, 2015:1821-1826.
[14] NG H W, WINKLER S. A data-driven approach to cleaning large face datasets[C]//Proceedings of the 2014 IEEE International Conference on Image Processing. Piscataway, NJ:IEEE, 2014:343-347.
[15] PEARSON K. Notes on the history of correlation[J]. Biometrika, 1920, 13(1):25-45.
[16] WANG B B, HAO X J, CHEN L S, et al. Face recognition based on the feature fusion of 2DLDA and LBP[C]//Proceedings of the 20134th IEEE International Conference on Information, Intelligence, Systems, and Applications. Piscataway, NJ:IEEE, 2013:1-6.
[17] BOODOO-JAHANGEER N B, BAICHOO S. Fusion of local and global features for face recognition[C]//Proceedings of the 2015 IEEE International Conference on Computing, Communication and Security. Piscataway, NJ:IEEE, 2015:1-8.
[18] CAMENT L A, CASTILLO L E, PEREZ J P, et al. Fusion of local normalization and Gabor entropy weighted features for face identification[J]. Pattern Recognition, 2014, 47(2):568-577.
[19] HAGHIGHAT M, ABDEL-MOTTALEB M, ALHALABI W. Fully automatic face normalization and single sample face recognition in unconstrained environments[J]. Expert Systems with Applications, 2016, 47:23-34.
[20] SZABO N S, TANAKA R I. Residue arithmetic and its applications to computer technology[M]. New York:McGraw-Hill Book Company, 1967:143-150.
[21] PESARAN H H, SMITH R J, et al. A generalized R^2 criterion for regression models estimated by the instrumental variables method[J]. Econometrica, 1994, 62(3):705-710.
[22] LIU S, FAN Y Y, GUO Z, et al. 2.5D facial attractiveness computation based on data-driven geometric ratios[C]//Proceedings of the 2015 International Conference on Intelligent Science and Big Data Engineering, LNCS 9242. Berlin:Springer, 2015:564-573.
[23] WANG S Y, SHAO M, FU Y. Attractive or Not?:Beauty prediction with attractiveness-aware encoders and robust late fusion[C]//Proceedings of the 22nd ACM International Conference on Multimedia. New York:ACM, 2014:805-808.
[24] XU J, JIN L W, LIANG L Y, et al. A new humanlike facial attractiveness predictor with cascaded fine-tuning deep learning model[J]. arXiv Preprint, 2015:arXiv.1511.02465.
[25] WHITEHILL J, MOVELLAN J R. Personalized facial attractiveness prediction[C]//Proceedings of the 8th IEEE International Conference on Automatic Face & Gesture Recognition. Piscataway, NJ:IEEE, 2008:1-7.
[26] ROTHE R, TIMOFTE R, van GOOL L. Some like it hot-visual guidance for preference prediction[C]//Proceedings of the 2016 IEEE Computer Vision and Pattern Recognition. Piscataway, NJ:IEEE:2016:5553-5561.

基于特征级和决策级融合的人脸吸引力评价方法

Facial attractiveness evaluation method based on fusion of feature-level and decision-level

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 3

编辑推荐

Metrics

[1]	何刚, 霍宏, 方涛. 兼顾特征级和决策级融合的场景分类[J]. 计算机应用, 2016, 36(5): 1262-1266.
[2]	陆惠玲, 周涛, 王惠群, 王文文. 基于特征级融合神经网络的磁共振成像前列腺肿瘤CAD模型[J]. 计算机应用, 2015, 35(10): 2813-2818.
[3]	李博曹鹏栗伟赵大哲. 基于尺度空间中多特征融合的医学影像分类[J]. 计算机应用, 2013, 33(04): 1108-1111.