Joint Feature and Similarity Deep Learning for Vehicle Re-identification
In this paper, a joint feature and similarity deep learning (JFSDL) method for vehicle reidentification is proposed. The proposed JFSDL method applies a siamese deep network to extract deep learning features for an input vehicle image pair simultaneously. The siamese deep network is learned under th...
Main Authors: | , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2018-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/8424333/ |
id |
doaj-5a775f538cbb4df6b6fd48196daacea2 |
---|---|
record_format |
Article |
spelling |
doaj-5a775f538cbb4df6b6fd48196daacea22021-03-29T20:51:03ZengIEEEIEEE Access2169-35362018-01-016437244373110.1109/ACCESS.2018.28623828424333Joint Feature and Similarity Deep Learning for Vehicle Re-identificationJianqing Zhu0https://orcid.org/0000-0001-8840-3629Huanqiang Zeng1Yongzhao Du2Zhen Lei3Lixin Zheng4https://orcid.org/0000-0002-5146-8661Canhui Cai5Fujian Provincial Academic Engineering Research Centre in Industrial Intellectual Techniques and Systems, College of Engineering, Huaqiao University, Quanzhou, ChinaCollege of Information Science and Engineering, Huaqiao University, Xiamen, ChinaFujian Provincial Academic Engineering Research Centre in Industrial Intellectual Techniques and Systems, College of Engineering, Huaqiao University, Quanzhou, ChinaCenter for Biometrics and Security Research and the National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences, Beijing, ChinaFujian Provincial Academic Engineering Research Centre in Industrial Intellectual Techniques and Systems, College of Engineering, Huaqiao University, Quanzhou, ChinaFujian Provincial Academic Engineering Research Centre in Industrial Intellectual Techniques and Systems, College of Engineering, Huaqiao University, Quanzhou, ChinaIn this paper, a joint feature and similarity deep learning (JFSDL) method for vehicle reidentification is proposed. The proposed JFSDL method applies a siamese deep network to extract deep learning features for an input vehicle image pair simultaneously. The siamese deep network is learned under the joint identification and verification supervision. The joint identification and verification supervision is realized by linearly combining two softmax functions and one hybrid similarity learning function. Moreover, based on the hybrid similarity learning function, the similarity score between the input vehicle image pair is also obtained by simultaneously projecting the element-wise absolute difference and multiplication of the corresponding deep learning feature pair with a group of learned weight coefficients. Extensive experiments show that the proposed JFSDL method is superior to multiple state-of-the-art vehicle re-identification methods on both the VehicleID and VeRi data sets.https://ieeexplore.ieee.org/document/8424333/Vehicle re-identificationfeature representationsimilarity learningdeep learning |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Jianqing Zhu Huanqiang Zeng Yongzhao Du Zhen Lei Lixin Zheng Canhui Cai |
spellingShingle |
Jianqing Zhu Huanqiang Zeng Yongzhao Du Zhen Lei Lixin Zheng Canhui Cai Joint Feature and Similarity Deep Learning for Vehicle Re-identification IEEE Access Vehicle re-identification feature representation similarity learning deep learning |
author_facet |
Jianqing Zhu Huanqiang Zeng Yongzhao Du Zhen Lei Lixin Zheng Canhui Cai |
author_sort |
Jianqing Zhu |
title |
Joint Feature and Similarity Deep Learning for Vehicle Re-identification |
title_short |
Joint Feature and Similarity Deep Learning for Vehicle Re-identification |
title_full |
Joint Feature and Similarity Deep Learning for Vehicle Re-identification |
title_fullStr |
Joint Feature and Similarity Deep Learning for Vehicle Re-identification |
title_full_unstemmed |
Joint Feature and Similarity Deep Learning for Vehicle Re-identification |
title_sort |
joint feature and similarity deep learning for vehicle re-identification |
publisher |
IEEE |
series |
IEEE Access |
issn |
2169-3536 |
publishDate |
2018-01-01 |
description |
In this paper, a joint feature and similarity deep learning (JFSDL) method for vehicle reidentification is proposed. The proposed JFSDL method applies a siamese deep network to extract deep learning features for an input vehicle image pair simultaneously. The siamese deep network is learned under the joint identification and verification supervision. The joint identification and verification supervision is realized by linearly combining two softmax functions and one hybrid similarity learning function. Moreover, based on the hybrid similarity learning function, the similarity score between the input vehicle image pair is also obtained by simultaneously projecting the element-wise absolute difference and multiplication of the corresponding deep learning feature pair with a group of learned weight coefficients. Extensive experiments show that the proposed JFSDL method is superior to multiple state-of-the-art vehicle re-identification methods on both the VehicleID and VeRi data sets. |
topic |
Vehicle re-identification feature representation similarity learning deep learning |
url |
https://ieeexplore.ieee.org/document/8424333/ |
work_keys_str_mv |
AT jianqingzhu jointfeatureandsimilaritydeeplearningforvehiclereidentification AT huanqiangzeng jointfeatureandsimilaritydeeplearningforvehiclereidentification AT yongzhaodu jointfeatureandsimilaritydeeplearningforvehiclereidentification AT zhenlei jointfeatureandsimilaritydeeplearningforvehiclereidentification AT lixinzheng jointfeatureandsimilaritydeeplearningforvehiclereidentification AT canhuicai jointfeatureandsimilaritydeeplearningforvehiclereidentification |
_version_ |
1724194041926516736 |