Research on Real-Time Face Recognition Algorithm Based on Lightweight Network
In order to achieve high-precision real-time face recognition on embedded and mobile devices, the advant-ages and disadvantages of common networks in face recognition are analyzed, and an efficient deep convolution neural network model Lightfacenet is proposed. In the network, a lightweight neural n...
Main Author: | |
---|---|
Format: | Article |
Language: | zho |
Published: |
Journal of Computer Engineering and Applications Beijing Co., Ltd., Science Press
2020-02-01
|
Series: | Jisuanji kexue yu tansuo |
Subjects: | |
Online Access: | http://fcst.ceaj.org/CN/abstract/abstract2121.shtml |
id |
doaj-d045f19629c247a799f1c558cfcda6ad |
---|---|
record_format |
Article |
spelling |
doaj-d045f19629c247a799f1c558cfcda6ad2021-08-09T10:20:28ZzhoJournal of Computer Engineering and Applications Beijing Co., Ltd., Science PressJisuanji kexue yu tansuo1673-94182020-02-0114231732410.3778/j.issn.1673-9418.1907037Research on Real-Time Face Recognition Algorithm Based on Lightweight NetworkZHANG Dian, WANG Haitao, JIANG Ying, CHEN Xing0Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming 650500, ChinaIn order to achieve high-precision real-time face recognition on embedded and mobile devices, the advant-ages and disadvantages of common networks in face recognition are analyzed, and an efficient deep convolution neural network model Lightfacenet is proposed. In the network, a lightweight neural network unit is proposed, which combines the deep separable convolution, point-by-point convolution, bottleneck structure and squeeze and excitation structure. The network can effectively solve the problem of parameter redundancy and large computation caused by the deep neural network with a certain accuracy, and then further improve the accuracy of the network through improved non-linear activation. The neural network not only retains some advantages of the convolutional neural network, but also balances the disadvantages of the network. In the same experimental environment, the Lightfacenet network not only achieves very high recognition accuracy, but also achieves real-time effect in the speed of model reasoning. After trained on MS-Celeb-1M data set, this model achieves 99.50% accuracy on LFW, and its effect is comparable to the advanced large deep convolutional neural network. For face recognition, Lightfacenet improves efficiency while ensuring accuracy compared to the most advanced mobile convolutional neural networks.http://fcst.ceaj.org/CN/abstract/abstract2121.shtmlface recognitionlightweight neural network unitreal timenon-linear activation |
collection |
DOAJ |
language |
zho |
format |
Article |
sources |
DOAJ |
author |
ZHANG Dian, WANG Haitao, JIANG Ying, CHEN Xing |
spellingShingle |
ZHANG Dian, WANG Haitao, JIANG Ying, CHEN Xing Research on Real-Time Face Recognition Algorithm Based on Lightweight Network Jisuanji kexue yu tansuo face recognition lightweight neural network unit real time non-linear activation |
author_facet |
ZHANG Dian, WANG Haitao, JIANG Ying, CHEN Xing |
author_sort |
ZHANG Dian, WANG Haitao, JIANG Ying, CHEN Xing |
title |
Research on Real-Time Face Recognition Algorithm Based on Lightweight Network |
title_short |
Research on Real-Time Face Recognition Algorithm Based on Lightweight Network |
title_full |
Research on Real-Time Face Recognition Algorithm Based on Lightweight Network |
title_fullStr |
Research on Real-Time Face Recognition Algorithm Based on Lightweight Network |
title_full_unstemmed |
Research on Real-Time Face Recognition Algorithm Based on Lightweight Network |
title_sort |
research on real-time face recognition algorithm based on lightweight network |
publisher |
Journal of Computer Engineering and Applications Beijing Co., Ltd., Science Press |
series |
Jisuanji kexue yu tansuo |
issn |
1673-9418 |
publishDate |
2020-02-01 |
description |
In order to achieve high-precision real-time face recognition on embedded and mobile devices, the advant-ages and disadvantages of common networks in face recognition are analyzed, and an efficient deep convolution neural network model Lightfacenet is proposed. In the network, a lightweight neural network unit is proposed, which combines the deep separable convolution, point-by-point convolution, bottleneck structure and squeeze and excitation structure. The network can effectively solve the problem of parameter redundancy and large computation caused by the deep neural network with a certain accuracy, and then further improve the accuracy of the network through improved non-linear activation. The neural network not only retains some advantages of the convolutional neural network, but also balances the disadvantages of the network. In the same experimental environment, the Lightfacenet network not only achieves very high recognition accuracy, but also achieves real-time effect in the speed of model reasoning. After trained on MS-Celeb-1M data set, this model achieves 99.50% accuracy on LFW, and its effect is comparable to the advanced large deep convolutional neural network. For face recognition, Lightfacenet improves efficiency while ensuring accuracy compared to the most advanced mobile convolutional neural networks. |
topic |
face recognition lightweight neural network unit real time non-linear activation |
url |
http://fcst.ceaj.org/CN/abstract/abstract2121.shtml |
work_keys_str_mv |
AT zhangdianwanghaitaojiangyingchenxing researchonrealtimefacerecognitionalgorithmbasedonlightweightnetwork |
_version_ |
1721214302407360512 |