基于改进Mobilenet算法的深度人脸识别算法
|
摘要
针对深度人脸识别任务在移动端遇到的存储空间受限、预测所需时间长、算法性能不高等问题,提出了一种改进的Mobilenet算法。将Mobilenet算法的监督信号Softmax改进为AM-Softmax,通过多次实验,设计出AMSoftmax比较适合Mobilenet算法的附加余量和缩放因子值。训练集和验证集来源于数据集MS-Celeb-1M-v1c和数据集Asian-Celeb,并在LFW数据集上对改进Mobilenet算法的有效性进行了验证。通过与初始Mobilenet算法模型的对比实验发现,采用改进Mobilenet算法的性能较优,准确率比softmax提升了十个百分点。充分利用数据集AsianCeleb中的亚洲名人ID,增加训练样本数,将性能进一步提高了四个百分点。
This paper proposes an improved Mobilenet algorithm for problems that the deep face recognition task encounters,such as limited storage space on the mobile terminal, long time required for prediction, and low performance of the algorithm.The supervised signal of Mobilenet algorithm is improved to AM-Softmax from Softmax. Through many experiments, AMSoftmax is designed to be more suitable for Mobilenet algorithm with suitable additional margin and scale value. The training set and verification set are derived from the dataset MS-Celeb-1M-v1 c and dataset Asian-Celeb, and the effectiveness of the improved Mobilenet algorithm is verified on the LFW data set. Compared with the initial Mobilenet algorithm model, it is found that the performance of the improved Mobilenet algorithm is better, and the accuracy rate is increased by 10% compared with softmax. By making full use of the Asian celebrity ID in the dataset Asian-Celeb to increase the number of training samples,performance is further improved by four percentage points.
This paper proposes an improved Mobilenet algorithm for problems that the deep face recognition task encounters,such as limited storage space on the mobile terminal, long time required for prediction, and low performance of the algorithm.The supervised signal of Mobilenet algorithm is improved to AM-Softmax from Softmax. Through many experiments, AMSoftmax is designed to be more suitable for Mobilenet algorithm with suitable additional margin and scale value. The training set and verification set are derived from the dataset MS-Celeb-1M-v1 c and dataset Asian-Celeb, and the effectiveness of the improved Mobilenet algorithm is verified on the LFW data set. Compared with the initial Mobilenet algorithm model, it is found that the performance of the improved Mobilenet algorithm is better, and the accuracy rate is increased by 10% compared with softmax. By making full use of the Asian celebrity ID in the dataset Asian-Celeb to increase the number of training samples,performance is further improved by four percentage points.
引文
[1]Taigman Y,Yang M,Ranzato M A,et al.Deepface:Closing the gap to human-level performance in face verification[C]//the IEEE Conference on Computer Vision and Pattern Recognition,2014:1701-1708
[2]Sun Yi,Wang Xiaogang,Tang Xiaoou.Deep learning face representation from predicting 10 000 classes[C]//the IEEE Conference on Computer Vision and Pattern Recognition,2014:1891-1898
[3]Schroff F,Kalenichenko D,Philbin J.Facenet:A unified embedding for face recognition and clustering[C]//the IEEE Conference on Computer Vision and Pattern Recognition,2015:815-823
[4]景晨凯,宋涛,庄雷,等.基于深度卷积神经网络的人脸识别技术综述[J].计算机应用与软件,2018,35(1):223-231
[5]Kemelmacher-shlizerman I,Seitz S M,Miller D,et al.The MegaFace Benchmark:1 Million Faces for Recognition at Scale[C]//Computer Vision and Pattern Recognition.IEEE,2016:4873-4882
[6]Schroff F,Kalenichenko D,Philbin J.Facenet:A unified embedding for face recognition and clustering[EB/OL].[2019-01-10].https://arxiv.org/abs/1503.03832
[7]Liu W,Wen Y,Yu Z,et al.Large-margin softmax loss for convolutional neural networks[EB/OL].[2019-01-10].https://arxiv.org/abs/1612.02295
[8]Wen Y,Zhang K,Li Z,et al.A Discriminative Feature Learning Approach for Deep Face Recognition[M]//Computer Vision-ECCV 2016.Springer Internationa Publishing,2016
[9]Liu Weiyang,Wen Yandong,Yu Zhiding,et al.SphereFace:Deep Hypersphere Embedding for Face Recognition[EB/OL].[2019-01-10].https://arxiv.org/abs/1704.08063
[10]Deng J,Guo J,Zafeiriou S.ArcFace:Additive Angular Margin Loss for Deep Face Recognition[EB/OL].[2019-01-10].https://arxiv.org/abs/1801.07698v1
[11]何长婷.课堂签到系统中的人脸识别方法研究与实现[D].安徽:中国科技大学,2018
[12]Iandola F N,Han S.Moskevvicz M W,et al.SqueezeNet:AlexNet-level accuracy with 50x fewer parameters and<0.5 MB model size[EB/OL].[2019-01-10].https://arxiv.org/abs/1602.07360
[13]Zhang X,Zhou X,Lin-M,et al.Shufilenet:An extremely efficient convolutional neural network for mobile devices[E B/O L].[2019-01-10].https://arxiv.org/abs/1707.01083
[14]Howard A G,Zhu M,Chen B,et al.Mobilenets:Efficient convolutional neural networks for mobile vision applications[EB/OL].[2019-01-10].https://arxiv.org/abs/1704.04861
[15]张萌.基于Boosting人脸检测算法的设计与实现[D].南京:东南大学,2016
[16]Huang G B,Ramesh M,Berg T,et al.Labeled faces in the wild:A database for studying face recognition in unconstrained environments[EB/OL].[2019-01-10].https://www.researchgate.net/publication/29622837_Labeled_Faces_in_the_Wild_A_Database_forStudying_Face_Recognition_in_Unconstrained_Environments
[2]Sun Yi,Wang Xiaogang,Tang Xiaoou.Deep learning face representation from predicting 10 000 classes[C]//the IEEE Conference on Computer Vision and Pattern Recognition,2014:1891-1898
[3]Schroff F,Kalenichenko D,Philbin J.Facenet:A unified embedding for face recognition and clustering[C]//the IEEE Conference on Computer Vision and Pattern Recognition,2015:815-823
[4]景晨凯,宋涛,庄雷,等.基于深度卷积神经网络的人脸识别技术综述[J].计算机应用与软件,2018,35(1):223-231
[5]Kemelmacher-shlizerman I,Seitz S M,Miller D,et al.The MegaFace Benchmark:1 Million Faces for Recognition at Scale[C]//Computer Vision and Pattern Recognition.IEEE,2016:4873-4882
[6]Schroff F,Kalenichenko D,Philbin J.Facenet:A unified embedding for face recognition and clustering[EB/OL].[2019-01-10].https://arxiv.org/abs/1503.03832
[7]Liu W,Wen Y,Yu Z,et al.Large-margin softmax loss for convolutional neural networks[EB/OL].[2019-01-10].https://arxiv.org/abs/1612.02295
[8]Wen Y,Zhang K,Li Z,et al.A Discriminative Feature Learning Approach for Deep Face Recognition[M]//Computer Vision-ECCV 2016.Springer Internationa Publishing,2016
[9]Liu Weiyang,Wen Yandong,Yu Zhiding,et al.SphereFace:Deep Hypersphere Embedding for Face Recognition[EB/OL].[2019-01-10].https://arxiv.org/abs/1704.08063
[10]Deng J,Guo J,Zafeiriou S.ArcFace:Additive Angular Margin Loss for Deep Face Recognition[EB/OL].[2019-01-10].https://arxiv.org/abs/1801.07698v1
[11]何长婷.课堂签到系统中的人脸识别方法研究与实现[D].安徽:中国科技大学,2018
[12]Iandola F N,Han S.Moskevvicz M W,et al.SqueezeNet:AlexNet-level accuracy with 50x fewer parameters and<0.5 MB model size[EB/OL].[2019-01-10].https://arxiv.org/abs/1602.07360
[13]Zhang X,Zhou X,Lin-M,et al.Shufilenet:An extremely efficient convolutional neural network for mobile devices[E B/O L].[2019-01-10].https://arxiv.org/abs/1707.01083
[14]Howard A G,Zhu M,Chen B,et al.Mobilenets:Efficient convolutional neural networks for mobile vision applications[EB/OL].[2019-01-10].https://arxiv.org/abs/1704.04861
[15]张萌.基于Boosting人脸检测算法的设计与实现[D].南京:东南大学,2016
[16]Huang G B,Ramesh M,Berg T,et al.Labeled faces in the wild:A database for studying face recognition in unconstrained environments[EB/OL].[2019-01-10].https://www.researchgate.net/publication/29622837_Labeled_Faces_in_the_Wild_A_Database_forStudying_Face_Recognition_in_Unconstrained_Environments