基于ThetaMEX全局池化的人脸识别神经网络——ShuffaceNet  

ShuffaceNet:face recognition neural network based on ThetaMEX global pooling

作　　者：陈侃松 郑园 许立君 王周宇 张哲 姚福娟 CHEN Kansong;ZHENG Yuan;XU Lijun;WANG Zhouyu;ZHANG Zhe;YAO Fujuan(School of Computer Science and Information Engineering,Hubei University,Wuhan Hubei 430062,China)

机构地区：[1]湖北大学计算机与信息工程学院,武汉430062

出　　处：《计算机应用》2023年第8期2572-2580,共9页journal of Computer Applications

基　　金：湖北省科技重大专项(202011901203001);湖北省重点研发计划项目(2021BAA184,2022BAA045);武汉市知识创新专项-曙光计划项目(2022010801020327)。

摘　　要：针对目前大规模网络不适合在手机、平板电脑等资源匮乏的移动设备上使用,以及池化层会导致特征图的稀疏性最终影响神经网络识别精度的问题,提出了一个轻量级人脸识别神经网络ShuffaceNet,设计了一个非线性平滑Log-Mean-Exp函数ThetaMEX,并提出了一种端到端可训练的ThetaMEX全局池化层(TGPL),从而在保证算法精度的前提下,减少网络参数、提高运算速度,进而达到有效地将该网络部署在资源匮乏的移动设备上的目的。ShuffaceNet约有3 600个参数,模型大小仅为3.5 MB。在LFW(Labled Faces in the Wild)、AgeDB-30 (Age Database-30)、CFP (Celebrities in Frontal Profile)人脸数据集上的识别测试的结果表明,ShuffaceNet的精度分别达到了99.32%、93.17%、94.51%。与MobileNetV1、SqueezeNet、Xception相比,所提网络的大小分别缩减了73.1%、82.1%、78.5%,在AgeDB-30数据集上的精度分别提高了5.0%、6.3%、6.7%。可见,基于ThetaMEX全局池化的所提网络能够提高模型精度。Focused on the issue that the current large-scale networks are not suitable to be applied on resource-starved mobile devices like smart phones and tablet computers,and the pooling layer will lead to the sparsity of feature map,which ultimately affect the recognition accuracy of the neural network,a lightweight face recognition neural network namely ShuffaceNet was proposed,a smooth nonlinear Log-Mean-Exp function ThetaMEX was designed,and an end-to-end trainable ThetaMEX Global Pool Layer(TGPL) was proposed,so as to reduce network parameters and improve computing speed while ensuring the accuracy of the algorithm,achieving the purpose that the network can be effectively deployed on mobile devices with limited resources.ShuffaceNet has about 3 600 parameters,and the model size is only 3.5 MB.The recognition test results on LFW(Labled Faces in the Wild),AgeDB-30(Age Database-30) and CFP(Celebrities in Frontal Profile) face datasets show that the accuracy of ShuffaceNet reaches 99.32%,93.17%,94.51% respectively.Compared with the traditional networks such as MobileNetV1,SqueezeNet and Xception,the proposed network has the size reduced by 73.1%,82.1% and 78.5% respectively,and the accuracy on AgeDB-30 dataset improved by 5.0%,6.3% and 6.7%respectively.It can be seen that the proposed network based on ThetaMEX global pooling can improve the model accuracy.

关 键 词：人脸识别 智能全局池化 ThetaMEX 神经网络 轻量级模型 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]