Lightweight FaceNet Based on MobileNet

• 作者: Xinzheng Xu Meng Du Huanxiu Guo Jianying Chang Xiaoyang Zhao
• 来源期刊: 智能科学国际期刊(英文)
• 学科: 数学
• 关键词: Face Recognition Deep Learning FaceNet MobileNet
• 年，卷（期）: 2021,（1）
• 页码范围: 1-16
• 页数: 16页
• 分类号: O17

摘要

Face recognition is a kind of biometric technology that recognizes identities through human faces. At first, the speed of machine recognition of human faces was slow and the accuracy was lower than manual recognition. With the rapid development of deep learning and the application of Convolutional Neural Network (CNN) in the field of face recognition, the accuracy of face recognition has greatly improved. FaceNet is a deep learning framework commo</span><span><span style="font-family:Verdana;">nly used in face recognition in recent years. FaceNet uses the deep learning model GoogLeNet, which has </span><span style="font-family:Verdana;">a high</span><span style="font-family:Verdana;"> accuracy in face recognition. However, its network structure is too large, which causes the </span><span style="font-family:Verdana;">FaceNet</span><span style="font-family:Verdana;"> to run at a low speed. Therefore, to improve the running speed without affecting the recognition accuracy of FaceNet, this paper proposes a lightweight FaceNet model based on MobileNet. This article mainly does the following works:</span></span></span><span style="font-family:""> </span><span style="font-family:Verdana;">Based on the analysis of the low running speed of FaceNet and the principle of MobileNet, a lightweight FaceNet model based on MobileNet is proposed. The model would reduce the overall calculation of the network by using deep separable convolutio</span><span style="font-family:""><span style="font-family:Verdana;">ns. In this paper, the model is trained on the CASIA-WebFace and VGGFace2 </span><span style="font-family:Verdana;">datasets,</span><span style="font-family:Verdana;"> and tested on the LFW dataset. Experimental results show that the model reduces the network parameters to a large extent while ensuring </span><span style="font-family:Verdana;">the accuracy</span><span style="font-family:Verdana;"> and hence an increase in system computing speed. The model can also perform face recognition on a specific person in the video.