Eintrag weiter verarbeiten
author_facet |
Chang, Shuo Li, Wei Zhang, Yifan Feng, Zhiyong Chang, Shuo Li, Wei Zhang, Yifan Feng, Zhiyong |
---|---|
author |
Chang, Shuo Li, Wei Zhang, Yifan Feng, Zhiyong |
spellingShingle |
Chang, Shuo Li, Wei Zhang, Yifan Feng, Zhiyong Sensors Online Siamese Network for Visual Object Tracking Electrical and Electronic Engineering Biochemistry Instrumentation Atomic and Molecular Physics, and Optics Analytical Chemistry |
author_sort |
chang, shuo |
spelling |
Chang, Shuo Li, Wei Zhang, Yifan Feng, Zhiyong 1424-8220 MDPI AG Electrical and Electronic Engineering Biochemistry Instrumentation Atomic and Molecular Physics, and Optics Analytical Chemistry http://dx.doi.org/10.3390/s19081858 <jats:p>Offline-trained Siamese networks are not robust to the environmental complication in visual object tracking. Without online learning, the Siamese network cannot learn from instance domain knowledge and adapt to appearance changes of targets. In this paper, a new lightweight Siamese network is proposed for feature extraction. To cope with the dynamics of targets and backgrounds, the weight in the proposed Siamese network is updated in an online manner during the tracking process. In order to enhance the discrimination capability, the cross-entropy loss is integrated into the contrastive loss. Inspired by the face verification algorithm DeepID2, the Bayesian verification model is applied for candidate selection. In general, visual object tracking can benefit from face verification algorithms. Numerical results suggest that the newly developed algorithm achieves comparable performance in public benchmarks.</jats:p> Online Siamese Network for Visual Object Tracking Sensors |
doi_str_mv |
10.3390/s19081858 |
facet_avail |
Online Free |
finc_class_facet |
Technik Mathematik Physik Chemie und Pharmazie Allgemeines |
format |
ElectronicArticle |
fullrecord |
blob:ai-49-aHR0cDovL2R4LmRvaS5vcmcvMTAuMzM5MC9zMTkwODE4NTg |
id |
ai-49-aHR0cDovL2R4LmRvaS5vcmcvMTAuMzM5MC9zMTkwODE4NTg |
institution |
DE-Zwi2 DE-D161 DE-Gla1 DE-Zi4 DE-15 DE-Pl11 DE-Rs1 DE-105 DE-14 DE-Ch1 DE-L229 DE-D275 DE-Bn3 DE-Brt1 |
imprint |
MDPI AG, 2019 |
imprint_str_mv |
MDPI AG, 2019 |
issn |
1424-8220 |
issn_str_mv |
1424-8220 |
language |
English |
mega_collection |
MDPI AG (CrossRef) |
match_str |
chang2019onlinesiamesenetworkforvisualobjecttracking |
publishDateSort |
2019 |
publisher |
MDPI AG |
recordtype |
ai |
record_format |
ai |
series |
Sensors |
source_id |
49 |
title |
Online Siamese Network for Visual Object Tracking |
title_unstemmed |
Online Siamese Network for Visual Object Tracking |
title_full |
Online Siamese Network for Visual Object Tracking |
title_fullStr |
Online Siamese Network for Visual Object Tracking |
title_full_unstemmed |
Online Siamese Network for Visual Object Tracking |
title_short |
Online Siamese Network for Visual Object Tracking |
title_sort |
online siamese network for visual object tracking |
topic |
Electrical and Electronic Engineering Biochemistry Instrumentation Atomic and Molecular Physics, and Optics Analytical Chemistry |
url |
http://dx.doi.org/10.3390/s19081858 |
publishDate |
2019 |
physical |
1858 |
description |
<jats:p>Offline-trained Siamese networks are not robust to the environmental complication in visual object tracking. Without online learning, the Siamese network cannot learn from instance domain knowledge and adapt to appearance changes of targets. In this paper, a new lightweight Siamese network is proposed for feature extraction. To cope with the dynamics of targets and backgrounds, the weight in the proposed Siamese network is updated in an online manner during the tracking process. In order to enhance the discrimination capability, the cross-entropy loss is integrated into the contrastive loss. Inspired by the face verification algorithm DeepID2, the Bayesian verification model is applied for candidate selection. In general, visual object tracking can benefit from face verification algorithms. Numerical results suggest that the newly developed algorithm achieves comparable performance in public benchmarks.</jats:p> |
container_issue |
8 |
container_start_page |
0 |
container_title |
Sensors |
container_volume |
19 |
format_de105 |
Article, E-Article |
format_de14 |
Article, E-Article |
format_de15 |
Article, E-Article |
format_de520 |
Article, E-Article |
format_de540 |
Article, E-Article |
format_dech1 |
Article, E-Article |
format_ded117 |
Article, E-Article |
format_degla1 |
E-Article |
format_del152 |
Buch |
format_del189 |
Article, E-Article |
format_dezi4 |
Article |
format_dezwi2 |
Article, E-Article |
format_finc |
Article, E-Article |
format_nrw |
Article, E-Article |
_version_ |
1792341783979491328 |
geogr_code |
not assigned |
last_indexed |
2024-03-01T16:25:25.166Z |
geogr_code_person |
not assigned |
openURL |
url_ver=Z39.88-2004&ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fvufind.svn.sourceforge.net%3Agenerator&rft.title=Online+Siamese+Network+for+Visual+Object+Tracking&rft.date=2019-04-18&genre=article&issn=1424-8220&volume=19&issue=8&pages=1858&jtitle=Sensors&atitle=Online+Siamese+Network+for+Visual+Object+Tracking&aulast=Feng&aufirst=Zhiyong&rft_id=info%3Adoi%2F10.3390%2Fs19081858&rft.language%5B0%5D=eng |
SOLR | |
_version_ | 1792341783979491328 |
author | Chang, Shuo, Li, Wei, Zhang, Yifan, Feng, Zhiyong |
author_facet | Chang, Shuo, Li, Wei, Zhang, Yifan, Feng, Zhiyong, Chang, Shuo, Li, Wei, Zhang, Yifan, Feng, Zhiyong |
author_sort | chang, shuo |
container_issue | 8 |
container_start_page | 0 |
container_title | Sensors |
container_volume | 19 |
description | <jats:p>Offline-trained Siamese networks are not robust to the environmental complication in visual object tracking. Without online learning, the Siamese network cannot learn from instance domain knowledge and adapt to appearance changes of targets. In this paper, a new lightweight Siamese network is proposed for feature extraction. To cope with the dynamics of targets and backgrounds, the weight in the proposed Siamese network is updated in an online manner during the tracking process. In order to enhance the discrimination capability, the cross-entropy loss is integrated into the contrastive loss. Inspired by the face verification algorithm DeepID2, the Bayesian verification model is applied for candidate selection. In general, visual object tracking can benefit from face verification algorithms. Numerical results suggest that the newly developed algorithm achieves comparable performance in public benchmarks.</jats:p> |
doi_str_mv | 10.3390/s19081858 |
facet_avail | Online, Free |
finc_class_facet | Technik, Mathematik, Physik, Chemie und Pharmazie, Allgemeines |
format | ElectronicArticle |
format_de105 | Article, E-Article |
format_de14 | Article, E-Article |
format_de15 | Article, E-Article |
format_de520 | Article, E-Article |
format_de540 | Article, E-Article |
format_dech1 | Article, E-Article |
format_ded117 | Article, E-Article |
format_degla1 | E-Article |
format_del152 | Buch |
format_del189 | Article, E-Article |
format_dezi4 | Article |
format_dezwi2 | Article, E-Article |
format_finc | Article, E-Article |
format_nrw | Article, E-Article |
geogr_code | not assigned |
geogr_code_person | not assigned |
id | ai-49-aHR0cDovL2R4LmRvaS5vcmcvMTAuMzM5MC9zMTkwODE4NTg |
imprint | MDPI AG, 2019 |
imprint_str_mv | MDPI AG, 2019 |
institution | DE-Zwi2, DE-D161, DE-Gla1, DE-Zi4, DE-15, DE-Pl11, DE-Rs1, DE-105, DE-14, DE-Ch1, DE-L229, DE-D275, DE-Bn3, DE-Brt1 |
issn | 1424-8220 |
issn_str_mv | 1424-8220 |
language | English |
last_indexed | 2024-03-01T16:25:25.166Z |
match_str | chang2019onlinesiamesenetworkforvisualobjecttracking |
mega_collection | MDPI AG (CrossRef) |
physical | 1858 |
publishDate | 2019 |
publishDateSort | 2019 |
publisher | MDPI AG |
record_format | ai |
recordtype | ai |
series | Sensors |
source_id | 49 |
spelling | Chang, Shuo Li, Wei Zhang, Yifan Feng, Zhiyong 1424-8220 MDPI AG Electrical and Electronic Engineering Biochemistry Instrumentation Atomic and Molecular Physics, and Optics Analytical Chemistry http://dx.doi.org/10.3390/s19081858 <jats:p>Offline-trained Siamese networks are not robust to the environmental complication in visual object tracking. Without online learning, the Siamese network cannot learn from instance domain knowledge and adapt to appearance changes of targets. In this paper, a new lightweight Siamese network is proposed for feature extraction. To cope with the dynamics of targets and backgrounds, the weight in the proposed Siamese network is updated in an online manner during the tracking process. In order to enhance the discrimination capability, the cross-entropy loss is integrated into the contrastive loss. Inspired by the face verification algorithm DeepID2, the Bayesian verification model is applied for candidate selection. In general, visual object tracking can benefit from face verification algorithms. Numerical results suggest that the newly developed algorithm achieves comparable performance in public benchmarks.</jats:p> Online Siamese Network for Visual Object Tracking Sensors |
spellingShingle | Chang, Shuo, Li, Wei, Zhang, Yifan, Feng, Zhiyong, Sensors, Online Siamese Network for Visual Object Tracking, Electrical and Electronic Engineering, Biochemistry, Instrumentation, Atomic and Molecular Physics, and Optics, Analytical Chemistry |
title | Online Siamese Network for Visual Object Tracking |
title_full | Online Siamese Network for Visual Object Tracking |
title_fullStr | Online Siamese Network for Visual Object Tracking |
title_full_unstemmed | Online Siamese Network for Visual Object Tracking |
title_short | Online Siamese Network for Visual Object Tracking |
title_sort | online siamese network for visual object tracking |
title_unstemmed | Online Siamese Network for Visual Object Tracking |
topic | Electrical and Electronic Engineering, Biochemistry, Instrumentation, Atomic and Molecular Physics, and Optics, Analytical Chemistry |
url | http://dx.doi.org/10.3390/s19081858 |