The Facial Beauty Prediction of the Model of Deep PCANet
-
摘要: 深度卷积神经网络（Deep Convolution Neural Network, DCNN）在人脸识别、图像分类和目标检测领域已取得较好效果，并得到广泛应用；但是，在人脸美丽预测中却存在拟合效果欠佳、网络训练难度大等问题。深度PCANet模型，将深度主元分析网络（Principal Component Analysis Network，PCANet）作为特征提取器；采用无监督预训练提取网络参数，具有网络训练时间短、图像特征提取快等特点，能有效避免DCNN存在的问题。为此，本文将深度PCANet引入人脸美丽预测，对训练集图像采用多尺度预处理，训练深度PCANet。该模型可提取人脸图像的结构性全局特征，采用特征增强方法可生成更具表征能力的特征；运用线性支持向量机（Support Vector Machine, SVM）和随机森林（Random Forest, RF）回归器进行训练和预测。基于SCUT-FBP人脸美丽数据库的实验结果表明，深度PCANet模型具有结构简单、特征提取快和无需网络调参优化等特点；选择合适的图像尺度与采用特征增强方法可提高人脸美丽评价结果，证明了所提方法的有效性和可行性。
-
关键词:
- 深度主元分析网络模型 /
- 线性支持向量机回归 /
- 随机森林回归 /
- 人脸美丽预测
Abstract: Deep convolution neural network(DCNN) has achieved good results in face recognition, image classification and object detection, and has been widely used. However, DCNN has some problems in the facial beauty prediction, such as bad fitting effect, hard training etc. Deep PCANet model, using Principal Component Analysis Network (PCANet) as feature extractor. The model can get parameters through unsupervised pre-training with the advantages of less time in training and faster speed in feature extraction, which can effectively avoid the problems of DCNN. For this reason, the deep PCANet is introduced into the facial beauty prediction, and the training set image adopts multi-scale preprocessing to train the deep PCANet. The model can extract the structural global features of the face image, and the feature enhancement method can generate features with more representative capabilities. Finally, linear Support Vector Machine (SVM) and Random Forest (RF) regressors were used for training and prediction. Experimental results for facial beauty prediction on SCUT-FBP database show that deep PCANet has advantages of simpler structure, faster feature extraction, and few parameter adjustment; choosing the right image scale and using feature enhancement methods can improve the face aesthetic evaluation results, which proves that the network is valid and feasible. -
-
[1] F. Chang, C.H. Chou. A Bi-prototype Theory of Facial Attractiveness[J]. Neural Computation, 2009, 21(3): 890-910. [2] J. Fan, K. P. Chau, X. Wan, et al. Prediction of facial attractiveness from facial proportions[J]. Pattern Recognition, 2012: 2326-2334. [3] W.C. Chiang, H.H. Lin, C.S. Huang, et al. The cluster assessment of facial attractiveness using fuzzy neural network classifier based on 3D Moiré features[J]. Pattern Recognition, 2014, 47(3): 1249-1260. [4] Y. Eisenthal, G. Dror, E. Ruppin. Facial attractiveness: beauty and the machine[J]. Neural Computation, 2006, 18 (1): 119-142. [5] H. Mao, L. Jin, M. Du. Automatic classification of Chinese female facial beauty using Support Vector Machine[C]. Proc. IEEE SMC, 2009: 4842-4846. [6] D. Zhang, Q. Zhao, F. Chen. Quantitative analysis of human facial beauty using geometric features[J]. Pattern Recognition, 2011, 44(4): 940-950. [7] Yan H. Cost-sensitive ordinal regression for fully automatic facial beauty assessment[J]. Neurocomputing, 2014, 129(5):334-342. [8] D. Gray, K. Yu, W. Xu, Y. Gong. Predicting Facial Beauty without Landmark[J]. Computer Vision -ECCV 2010. vol 6316: 434-447. [9] D. Xie, L. Liang, L. W. Jin, J. Xu. SCUT-FBP: A Benchmark Dataset for Facial Beauty Perception[C], in: IEEE International Conference on Systems, 2015: 1821-1826. [10] Gan J,Li L,Zhai Y, Liu Y. Deep self-taught learning for facial beauty prediction[J]. Neurocomputing 2014(144): 295-303. [11] J. Xu, L. W. Jin, L. Liang, Z. Feng, D. Xie. A new humanlike facial attractiveness predictor with cascaded fine-tuning deep learning model[J]. Computer Science, 2015, 70(1): 45-79. [12] Y. Taigman, M. Yang, M. Ranzato, and L. Wolf. Deep-face: Closing the gap to human-level performance in face verification[C]. In proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition, 2014: 1701-1708. [13] Chan Tsung-Han, Jia K, PCANet: a simple deep learning baseline for image classification? [J]. IEEE Transaction on Image Processing, 2015, 24 (12): 5017-5032. [14] D., K. Yu, W. Xu, Y. Gong, Predicting Facial Beauty without Landmark[C], Computer Vision ECCV 2010, 2010, pp. 434-447. [15] Meguid M K A E, Levine M D. Fully automated recognition of spontaneous facial s in videos using random forest classifiers[J]. IEEE Transactions on Affective Computing, 2014,5(2): 141-154. [16] Y. Sun, Y. Chen, X. Wang, and X. Tang. Deep convolutional network cascade for facial point detection[C]. In Proceedings of the 2013 IEEE Conference on Computer Vision and Pattern Recognition, 2013, 5: 3476-3483. [17] J.Y. Gan, Y.K. Zhai, B. Wang, Unconstrained Facial Beauty Prediction Based on Multi-scale K-means[J]. Chinese Journal of Electronics. 25(1) (2016) 1-9. -
期刊类型引用(8)
1. 王豪，廖云朋，彭宽，黄忠朝. 基于语义分割的单导心电图心拍分类研究. 生物医学工程研究. 2024(03): 207-213 . 
百度学术
2. 朱椰毅，姜博文，范骁，林浩立，韦强. 基于毫米波雷达的城市轨道交通乘客统计分析方法研究. 现代城市轨道交通. 2024(08): 116-124 . 百度学术
3. 刘喜洋，余建宇，陈威，谢煜晨. 基于FPGA的改进单比特接收机门限设计方法研究. 火控雷达技术. 2023(01): 82-86 . 百度学术
4. 郭田雨，严荣国，方旭晨，徐玉玲，陶争屹. 基于希尔伯特变换和自适应阈值的R波检测算法. 计算机与现代化. 2022(02): 114-119 . 百度学术
5. 徐原，王少娜. 外界干扰下体外自动除颤仪室颤信号高精度检测方法. 自动化与仪器仪表. 2022(11): 291-295 . 百度学术
6. 戴世杰，李哲明. 基于融合模糊聚类算法的异常心电多频段弱信号快速捕捉方法. 现代电子技术. 2022(21): 62-66 . 百度学术
7. 孙浩，吴世鹏. 一种基于恒虚警检测的无线电监测方法. 通信技术. 2021(03): 647-653 . 百度学术
8. 余顺，张平，王曾. 基于K-means算法的QRS波群检测研究. 安徽职业技术学院学报. 2021(01): 25-28 . 百度学术
其他类型引用(5)
计量
- 文章访问数: 76
- HTML全文浏览量: 5
- PDF下载量: 582
- 被引次数: 13
下载:
