Real‐time long‐term tracking with reliability assessment and object recovery
Abstract In recent years, many visual tracking algorithms based on discriminative correlation filters have been proposed and proved to be successful in short‐term tracking. However, most algorithms do not handle long‐term tracking well due to factors such as occlusion and deformation. The aim of thi...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Wiley
2021-03-01
|
Series: | IET Image Processing |
Subjects: | |
Online Access: | https://doi.org/10.1049/ipr2.12072 |
_version_ | 1798028032257032192 |
---|---|
author | Jun Liu Ping Ye Xingchen Zhang Gang Xiao |
author_facet | Jun Liu Ping Ye Xingchen Zhang Gang Xiao |
author_sort | Jun Liu |
collection | DOAJ |
description | Abstract In recent years, many visual tracking algorithms based on discriminative correlation filters have been proposed and proved to be successful in short‐term tracking. However, most algorithms do not handle long‐term tracking well due to factors such as occlusion and deformation. The aim of this paper is to propose a long‐term tracking method with reliability assessment and object recovery. First, the relationship between the overlap rate and the response value is extensively studied, and then from the perspective of the time axis, the tracking process is evaluated by the fluctuation trend of the continuous response value. In the object recovery mechanism, we propose to alternately use local search and global search to improve the efficiency of detection. To this end, a sliding window is designed for cyclic shifting in the local region to achieve dense sampling within the region of interest, and EdgeBox is used in the global search to achieve target detection. Further, flexible switching between local search and global search is achieved by the difference in displacement of the object. Extensive experimental results on several benchmark datasets demonstrate that the proposed long‐term tracker can achieve state‐of‐the‐art accuracy with real‐time speed of about 27 frames per second. |
first_indexed | 2024-04-11T19:01:14Z |
format | Article |
id | doaj.art-e20d0ad7af594742a061175d20503a18 |
institution | Directory Open Access Journal |
issn | 1751-9659 1751-9667 |
language | English |
last_indexed | 2024-04-11T19:01:14Z |
publishDate | 2021-03-01 |
publisher | Wiley |
record_format | Article |
series | IET Image Processing |
spelling | doaj.art-e20d0ad7af594742a061175d20503a182022-12-22T04:08:02ZengWileyIET Image Processing1751-96591751-96672021-03-0115491893510.1049/ipr2.12072Real‐time long‐term tracking with reliability assessment and object recoveryJun Liu0Ping Ye1Xingchen Zhang2Gang Xiao3School of Aeronautics and Astronautics Shanghai Jiao Tong University ChinaSchool of Aeronautics and Astronautics Shanghai Jiao Tong University ChinaSchool of Aeronautics and Astronautics Shanghai Jiao Tong University ChinaSchool of Aeronautics and Astronautics Shanghai Jiao Tong University ChinaAbstract In recent years, many visual tracking algorithms based on discriminative correlation filters have been proposed and proved to be successful in short‐term tracking. However, most algorithms do not handle long‐term tracking well due to factors such as occlusion and deformation. The aim of this paper is to propose a long‐term tracking method with reliability assessment and object recovery. First, the relationship between the overlap rate and the response value is extensively studied, and then from the perspective of the time axis, the tracking process is evaluated by the fluctuation trend of the continuous response value. In the object recovery mechanism, we propose to alternately use local search and global search to improve the efficiency of detection. To this end, a sliding window is designed for cyclic shifting in the local region to achieve dense sampling within the region of interest, and EdgeBox is used in the global search to achieve target detection. Further, flexible switching between local search and global search is achieved by the difference in displacement of the object. Extensive experimental results on several benchmark datasets demonstrate that the proposed long‐term tracker can achieve state‐of‐the‐art accuracy with real‐time speed of about 27 frames per second.https://doi.org/10.1049/ipr2.12072Optical, image and video signal processingFiltering methods in signal processingComputer vision and image processing techniquesVideo signal processing |
spellingShingle | Jun Liu Ping Ye Xingchen Zhang Gang Xiao Real‐time long‐term tracking with reliability assessment and object recovery IET Image Processing Optical, image and video signal processing Filtering methods in signal processing Computer vision and image processing techniques Video signal processing |
title | Real‐time long‐term tracking with reliability assessment and object recovery |
title_full | Real‐time long‐term tracking with reliability assessment and object recovery |
title_fullStr | Real‐time long‐term tracking with reliability assessment and object recovery |
title_full_unstemmed | Real‐time long‐term tracking with reliability assessment and object recovery |
title_short | Real‐time long‐term tracking with reliability assessment and object recovery |
title_sort | real time long term tracking with reliability assessment and object recovery |
topic | Optical, image and video signal processing Filtering methods in signal processing Computer vision and image processing techniques Video signal processing |
url | https://doi.org/10.1049/ipr2.12072 |
work_keys_str_mv | AT junliu realtimelongtermtrackingwithreliabilityassessmentandobjectrecovery AT pingye realtimelongtermtrackingwithreliabilityassessmentandobjectrecovery AT xingchenzhang realtimelongtermtrackingwithreliabilityassessmentandobjectrecovery AT gangxiao realtimelongtermtrackingwithreliabilityassessmentandobjectrecovery |