A CPU Real-Time Face Alignment for Mobile Platform
Face alignment is a common technology in face recognition and face verification field. Previous works mostly pay attention to improving the accuracy of prediction and ignored the practicability of the method. In this paper, we aim at providing a two-stage face alignment network for mobile platform....
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2020-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/8952701/ |
id |
doaj-f8d03ee317304e8c8d4336b110b77f67 |
---|---|
record_format |
Article |
spelling |
doaj-f8d03ee317304e8c8d4336b110b77f672021-03-30T01:18:45ZengIEEEIEEE Access2169-35362020-01-0188834884310.1109/ACCESS.2020.29648388952701A CPU Real-Time Face Alignment for Mobile PlatformXin Ning0https://orcid.org/0000-0001-7897-1673Pengfei Duan1https://orcid.org/0000-0003-1637-259XWeijun Li2https://orcid.org/0000-0001-9668-2883Yuan Shi3https://orcid.org/0000-0001-5685-9437Shuang Li4https://orcid.org/0000-0002-3089-7221Institute of Semiconductors, Chinese Academy of Sciences, Beijing, ChinaCognitive Computing Technology Joint Laboratory, Wave Group, Beijing, ChinaInstitute of Semiconductors, Chinese Academy of Sciences, Beijing, ChinaCognitive Computing Technology Joint Laboratory, Wave Group, Beijing, ChinaInstitute of Semiconductors, Chinese Academy of Sciences, Beijing, ChinaFace alignment is a common technology in face recognition and face verification field. Previous works mostly pay attention to improving the accuracy of prediction and ignored the practicability of the method. In this paper, we aim at providing a two-stage face alignment network for mobile platform. Firstly, the network was trained with residual label which is the difference between ground truth and mean shape. Secondly, the input data in the second stage is composed of the original data and generated heatmap which enriched the data types. Finally, a new loss function is used to enhance the convergence of local region. Experimental results show that our method not only provides high precision but also improve the real-time processing performance on the mobile platforms.https://ieeexplore.ieee.org/document/8952701/Residual labelheatmapglobal poolingloss function |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Xin Ning Pengfei Duan Weijun Li Yuan Shi Shuang Li |
spellingShingle |
Xin Ning Pengfei Duan Weijun Li Yuan Shi Shuang Li A CPU Real-Time Face Alignment for Mobile Platform IEEE Access Residual label heatmap global pooling loss function |
author_facet |
Xin Ning Pengfei Duan Weijun Li Yuan Shi Shuang Li |
author_sort |
Xin Ning |
title |
A CPU Real-Time Face Alignment for Mobile Platform |
title_short |
A CPU Real-Time Face Alignment for Mobile Platform |
title_full |
A CPU Real-Time Face Alignment for Mobile Platform |
title_fullStr |
A CPU Real-Time Face Alignment for Mobile Platform |
title_full_unstemmed |
A CPU Real-Time Face Alignment for Mobile Platform |
title_sort |
cpu real-time face alignment for mobile platform |
publisher |
IEEE |
series |
IEEE Access |
issn |
2169-3536 |
publishDate |
2020-01-01 |
description |
Face alignment is a common technology in face recognition and face verification field. Previous works mostly pay attention to improving the accuracy of prediction and ignored the practicability of the method. In this paper, we aim at providing a two-stage face alignment network for mobile platform. Firstly, the network was trained with residual label which is the difference between ground truth and mean shape. Secondly, the input data in the second stage is composed of the original data and generated heatmap which enriched the data types. Finally, a new loss function is used to enhance the convergence of local region. Experimental results show that our method not only provides high precision but also improve the real-time processing performance on the mobile platforms. |
topic |
Residual label heatmap global pooling loss function |
url |
https://ieeexplore.ieee.org/document/8952701/ |
work_keys_str_mv |
AT xinning acpurealtimefacealignmentformobileplatform AT pengfeiduan acpurealtimefacealignmentformobileplatform AT weijunli acpurealtimefacealignmentformobileplatform AT yuanshi acpurealtimefacealignmentformobileplatform AT shuangli acpurealtimefacealignmentformobileplatform AT xinning cpurealtimefacealignmentformobileplatform AT pengfeiduan cpurealtimefacealignmentformobileplatform AT weijunli cpurealtimefacealignmentformobileplatform AT yuanshi cpurealtimefacealignmentformobileplatform AT shuangli cpurealtimefacealignmentformobileplatform |
_version_ |
1724187315061915648 |