Cargando…

LPCF: Robust Correlation Tracking via Locality Preserving Tracking Validation

In visual tracking, the tracking model must be updated online, which often leads to undesired inclusion of corrupted training samples, and hence inducing tracking failure. We present a locality preserving correlation filter (LPCF) integrating a novel and generic decontamination approach, which mitig...

Descripción completa

Detalles Bibliográficos
Autores principales: Zhou, Yixuan, Zhang, Weimin, Shi, Yongliang, Wang, Ziyu, Li, Fangxing, Huang, Qiang
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2020
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7731162/
https://www.ncbi.nlm.nih.gov/pubmed/33266108
http://dx.doi.org/10.3390/s20236853
_version_ 1783621846324740096
author Zhou, Yixuan
Zhang, Weimin
Shi, Yongliang
Wang, Ziyu
Li, Fangxing
Huang, Qiang
author_facet Zhou, Yixuan
Zhang, Weimin
Shi, Yongliang
Wang, Ziyu
Li, Fangxing
Huang, Qiang
author_sort Zhou, Yixuan
collection PubMed
description In visual tracking, the tracking model must be updated online, which often leads to undesired inclusion of corrupted training samples, and hence inducing tracking failure. We present a locality preserving correlation filter (LPCF) integrating a novel and generic decontamination approach, which mitigates the model drift problem. Our decontamination approach maintains the local neighborhood feature points structures of the bounding box center. This proposed tracking-result validation approach models not only the spatial neighborhood relationship but also the topological structures of the bounding box center. Additionally, a closed-form solution to our approach is derived, which makes the tracking-result validation process could be accomplished in only milliseconds. Moreover, a dimensionality reduction strategy is introduced to improve the real-time performance of our translation estimation component. Comprehensive experiments are performed on OTB-2015, LASOT, TrackingNet. The experimental results show that our decontamination approach remarkably improves the overall performance by 6.2%, 12.6%, and 3%, meanwhile, our complete algorithm improves the baseline by 27.8%, 34.8%, and 15%. Finally, our tracker achieves the best performance among most existing decontamination trackers under the real-time requirement.
format Online
Article
Text
id pubmed-7731162
institution National Center for Biotechnology Information
language English
publishDate 2020
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-77311622020-12-12 LPCF: Robust Correlation Tracking via Locality Preserving Tracking Validation Zhou, Yixuan Zhang, Weimin Shi, Yongliang Wang, Ziyu Li, Fangxing Huang, Qiang Sensors (Basel) Article In visual tracking, the tracking model must be updated online, which often leads to undesired inclusion of corrupted training samples, and hence inducing tracking failure. We present a locality preserving correlation filter (LPCF) integrating a novel and generic decontamination approach, which mitigates the model drift problem. Our decontamination approach maintains the local neighborhood feature points structures of the bounding box center. This proposed tracking-result validation approach models not only the spatial neighborhood relationship but also the topological structures of the bounding box center. Additionally, a closed-form solution to our approach is derived, which makes the tracking-result validation process could be accomplished in only milliseconds. Moreover, a dimensionality reduction strategy is introduced to improve the real-time performance of our translation estimation component. Comprehensive experiments are performed on OTB-2015, LASOT, TrackingNet. The experimental results show that our decontamination approach remarkably improves the overall performance by 6.2%, 12.6%, and 3%, meanwhile, our complete algorithm improves the baseline by 27.8%, 34.8%, and 15%. Finally, our tracker achieves the best performance among most existing decontamination trackers under the real-time requirement. MDPI 2020-11-30 /pmc/articles/PMC7731162/ /pubmed/33266108 http://dx.doi.org/10.3390/s20236853 Text en © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Zhou, Yixuan
Zhang, Weimin
Shi, Yongliang
Wang, Ziyu
Li, Fangxing
Huang, Qiang
LPCF: Robust Correlation Tracking via Locality Preserving Tracking Validation
title LPCF: Robust Correlation Tracking via Locality Preserving Tracking Validation
title_full LPCF: Robust Correlation Tracking via Locality Preserving Tracking Validation
title_fullStr LPCF: Robust Correlation Tracking via Locality Preserving Tracking Validation
title_full_unstemmed LPCF: Robust Correlation Tracking via Locality Preserving Tracking Validation
title_short LPCF: Robust Correlation Tracking via Locality Preserving Tracking Validation
title_sort lpcf: robust correlation tracking via locality preserving tracking validation
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7731162/
https://www.ncbi.nlm.nih.gov/pubmed/33266108
http://dx.doi.org/10.3390/s20236853
work_keys_str_mv AT zhouyixuan lpcfrobustcorrelationtrackingvialocalitypreservingtrackingvalidation
AT zhangweimin lpcfrobustcorrelationtrackingvialocalitypreservingtrackingvalidation
AT shiyongliang lpcfrobustcorrelationtrackingvialocalitypreservingtrackingvalidation
AT wangziyu lpcfrobustcorrelationtrackingvialocalitypreservingtrackingvalidation
AT lifangxing lpcfrobustcorrelationtrackingvialocalitypreservingtrackingvalidation
AT huangqiang lpcfrobustcorrelationtrackingvialocalitypreservingtrackingvalidation