IoT 18: e1

Research Article

A Lightweight Face Recognition Model based on MobileFaceNet for Limited Computation Environment

Download127 downloads
  • @ARTICLE{10.4108/eai.28-2-2022.173547,
        author={Jianyu Xiao and Guoli Jiang and Huanhua Liu},
        title={A Lightweight Face Recognition Model based on MobileFaceNet for Limited Computation Environment},
        journal={EAI Endorsed Transactions on Internet of Things: Online First},
        volume={},
        number={},
        publisher={EAI},
        journal_a={IOT},
        year={2022},
        month={2},
        keywords={Face recognition, MobileFaceNet, weak computing environment, channel attention mechanism},
        doi={10.4108/eai.28-2-2022.173547}
    }
    
  • Jianyu Xiao
    Guoli Jiang
    Huanhua Liu
    Year: 2022
    A Lightweight Face Recognition Model based on MobileFaceNet for Limited Computation Environment
    IOT
    EAI
    DOI: 10.4108/eai.28-2-2022.173547
Jianyu Xiao1, Guoli Jiang1, Huanhua Liu2,*
  • 1: School of Computer Science and Engineering, Central South University, Changsha 410075, China
  • 2: School of Information Technology and Management, Hunan University of Finance and Economics, Changsha 410205, China
*Contact email: liuhuanhua@hufe.edu.cn

Abstract

The face recognition method based on deep convolutional neural network is difficult to deploy in the embedding devices. In this work, we optimize the MobileFaceNet face recognition network MobileFaceNet so as to deploy it in embedding environment. Firstly, we reduce the model parameters by reducing the number of layers in MobileFaceNet. Then, the h-ReLU6 activation function is used to replace PReLU in the original model. Finally, the effective channel attention module efficient channel attention is introduced to obtain the importance of each feature channel by learning. After the optimization, the MobileFaceNet parameters are compressed to 3.4 MB, which is smaller than the original model (4.9 MB), and the mAPs reach 98.52%, 97.54% and 91.33% on the test sets of LFW, VGGFace2 and the self-built database, respectively, and the recognition time is about 85 ms/photo. It shows that the proposed method achieves a good balance between the model complexity and model performance.