No Cover Image

Journal article 387 views 143 downloads

Multi-Channel Features Spatio-Temporal Context Learning for Visual Tracking

Xiaoqin Zhou, Xiaofeng Liu, Chenguang Yang, Aimin Jiang, Bin Yan

IEEE Access, Volume: 5, Pages: 12856 - 12864

Swansea University Author: Chenguang Yang

Check full text

DOI (Published version): 10.1109/ACCESS.2017.2720746

Abstract

Visual tracking is a challenging issue in surveillance, human-computer interaction and intelligent robotics, among others. Managing appearance changes of the target object, illumination changes, rotations, non-rigid deformations, partial or full occlusions, background clutter, fast motion, and so fo...

Full description

Published in: IEEE Access
ISSN: 2169-3536
Published: 2017
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa34537
Tags: Add Tag
No Tags, Be the first to tag this record!
first_indexed 2017-07-02T05:18:22Z
last_indexed 2018-02-09T05:24:46Z
id cronfa34537
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2017-09-05T11:04:28.4057987</datestamp><bib-version>v2</bib-version><id>34537</id><entry>2017-07-02</entry><title>Multi-Channel Features Spatio-Temporal Context Learning for Visual Tracking</title><swanseaauthors><author><sid>d2a5024448bfac00a9b3890a8404380b</sid><ORCID/><firstname>Chenguang</firstname><surname>Yang</surname><name>Chenguang Yang</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2017-07-02</date><deptcode>EEN</deptcode><abstract>Visual tracking is a challenging issue in surveillance, human-computer interaction and intelligent robotics, among others. Managing appearance changes of the target object, illumination changes, rotations, non-rigid deformations, partial or full occlusions, background clutter, fast motion, and so forth is generally difficult. Among the numerous existing trackers, the correlationfilter- based tracker can achieve appealing performance with a fast speed for fast Fourier transform (FFT). Motivated by this property, the spatio-temporal context (STC) learning algorithm was proposed with consideration of the information from the context around the target, and this algorithm achieved good results. However, STC only utilizes the overall intensity information. In this paper, we propose a multi-channel features spatio-temporal context (MFSTC) learning algorithm with an improved scaleadaptive scheme. Our algorithm integrates powerful features, including Histogram of Oriented Gradients (HoG) and color naming, using kernel methods on the basis of the STC algorithm to further enhance the overall tracking performance. Extensive experimental results obtained from various benchmark datasets demonstrate that the proposed tracker is promising for various challenging scenarios and maintains real-time performance at an average speed of 78 fps. According to the test results, our algorithm outperforms the STC algorithm and achieves performance that is competitive with the state-of-the-art algorithms.</abstract><type>Journal Article</type><journal>IEEE Access</journal><volume>5</volume><paginationStart>12856</paginationStart><paginationEnd>12864</paginationEnd><publisher/><issnElectronic>2169-3536</issnElectronic><keywords/><publishedDay>30</publishedDay><publishedMonth>6</publishedMonth><publishedYear>2017</publishedYear><publishedDate>2017-06-30</publishedDate><doi>10.1109/ACCESS.2017.2720746</doi><url/><notes/><college>COLLEGE NANME</college><department>Engineering</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>EEN</DepartmentCode><institution>Swansea University</institution><apcterm/><lastEdited>2017-09-05T11:04:28.4057987</lastEdited><Created>2017-07-02T01:11:32.3041477</Created><path><level id="1">Faculty of Science and Engineering</level><level id="2">School of Engineering and Applied Sciences - Uncategorised</level></path><authors><author><firstname>Xiaoqin</firstname><surname>Zhou</surname><order>1</order></author><author><firstname>Xiaofeng</firstname><surname>Liu</surname><order>2</order></author><author><firstname>Chenguang</firstname><surname>Yang</surname><orcid/><order>3</order></author><author><firstname>Aimin</firstname><surname>Jiang</surname><order>4</order></author><author><firstname>Bin</firstname><surname>Yan</surname><order>5</order></author></authors><documents><document><filename>0034537-05092017110211.pdf</filename><originalFilename>ZhouMultiChannelFeatures.pdf</originalFilename><uploaded>2017-09-05T11:02:11.0630000</uploaded><type>Output</type><contentLength>4062890</contentLength><contentType>application/pdf</contentType><version>Version of Record</version><cronfaStatus>true</cronfaStatus><embargoDate>2017-09-05T00:00:00.0000000</embargoDate><copyrightCorrect>true</copyrightCorrect><language>eng</language></document></documents><OutputDurs/></rfc1807>
spelling 2017-09-05T11:04:28.4057987 v2 34537 2017-07-02 Multi-Channel Features Spatio-Temporal Context Learning for Visual Tracking d2a5024448bfac00a9b3890a8404380b Chenguang Yang Chenguang Yang true false 2017-07-02 EEN Visual tracking is a challenging issue in surveillance, human-computer interaction and intelligent robotics, among others. Managing appearance changes of the target object, illumination changes, rotations, non-rigid deformations, partial or full occlusions, background clutter, fast motion, and so forth is generally difficult. Among the numerous existing trackers, the correlationfilter- based tracker can achieve appealing performance with a fast speed for fast Fourier transform (FFT). Motivated by this property, the spatio-temporal context (STC) learning algorithm was proposed with consideration of the information from the context around the target, and this algorithm achieved good results. However, STC only utilizes the overall intensity information. In this paper, we propose a multi-channel features spatio-temporal context (MFSTC) learning algorithm with an improved scaleadaptive scheme. Our algorithm integrates powerful features, including Histogram of Oriented Gradients (HoG) and color naming, using kernel methods on the basis of the STC algorithm to further enhance the overall tracking performance. Extensive experimental results obtained from various benchmark datasets demonstrate that the proposed tracker is promising for various challenging scenarios and maintains real-time performance at an average speed of 78 fps. According to the test results, our algorithm outperforms the STC algorithm and achieves performance that is competitive with the state-of-the-art algorithms. Journal Article IEEE Access 5 12856 12864 2169-3536 30 6 2017 2017-06-30 10.1109/ACCESS.2017.2720746 COLLEGE NANME Engineering COLLEGE CODE EEN Swansea University 2017-09-05T11:04:28.4057987 2017-07-02T01:11:32.3041477 Faculty of Science and Engineering School of Engineering and Applied Sciences - Uncategorised Xiaoqin Zhou 1 Xiaofeng Liu 2 Chenguang Yang 3 Aimin Jiang 4 Bin Yan 5 0034537-05092017110211.pdf ZhouMultiChannelFeatures.pdf 2017-09-05T11:02:11.0630000 Output 4062890 application/pdf Version of Record true 2017-09-05T00:00:00.0000000 true eng
title Multi-Channel Features Spatio-Temporal Context Learning for Visual Tracking
spellingShingle Multi-Channel Features Spatio-Temporal Context Learning for Visual Tracking
Chenguang Yang
title_short Multi-Channel Features Spatio-Temporal Context Learning for Visual Tracking
title_full Multi-Channel Features Spatio-Temporal Context Learning for Visual Tracking
title_fullStr Multi-Channel Features Spatio-Temporal Context Learning for Visual Tracking
title_full_unstemmed Multi-Channel Features Spatio-Temporal Context Learning for Visual Tracking
title_sort Multi-Channel Features Spatio-Temporal Context Learning for Visual Tracking
author_id_str_mv d2a5024448bfac00a9b3890a8404380b
author_id_fullname_str_mv d2a5024448bfac00a9b3890a8404380b_***_Chenguang Yang
author Chenguang Yang
author2 Xiaoqin Zhou
Xiaofeng Liu
Chenguang Yang
Aimin Jiang
Bin Yan
format Journal article
container_title IEEE Access
container_volume 5
container_start_page 12856
publishDate 2017
institution Swansea University
issn 2169-3536
doi_str_mv 10.1109/ACCESS.2017.2720746
college_str Faculty of Science and Engineering
hierarchytype
hierarchy_top_id facultyofscienceandengineering
hierarchy_top_title Faculty of Science and Engineering
hierarchy_parent_id facultyofscienceandengineering
hierarchy_parent_title Faculty of Science and Engineering
department_str School of Engineering and Applied Sciences - Uncategorised{{{_:::_}}}Faculty of Science and Engineering{{{_:::_}}}School of Engineering and Applied Sciences - Uncategorised
document_store_str 1
active_str 0
description Visual tracking is a challenging issue in surveillance, human-computer interaction and intelligent robotics, among others. Managing appearance changes of the target object, illumination changes, rotations, non-rigid deformations, partial or full occlusions, background clutter, fast motion, and so forth is generally difficult. Among the numerous existing trackers, the correlationfilter- based tracker can achieve appealing performance with a fast speed for fast Fourier transform (FFT). Motivated by this property, the spatio-temporal context (STC) learning algorithm was proposed with consideration of the information from the context around the target, and this algorithm achieved good results. However, STC only utilizes the overall intensity information. In this paper, we propose a multi-channel features spatio-temporal context (MFSTC) learning algorithm with an improved scaleadaptive scheme. Our algorithm integrates powerful features, including Histogram of Oriented Gradients (HoG) and color naming, using kernel methods on the basis of the STC algorithm to further enhance the overall tracking performance. Extensive experimental results obtained from various benchmark datasets demonstrate that the proposed tracker is promising for various challenging scenarios and maintains real-time performance at an average speed of 78 fps. According to the test results, our algorithm outperforms the STC algorithm and achieves performance that is competitive with the state-of-the-art algorithms.
published_date 2017-06-30T03:42:52Z
_version_ 1763751988672069632
score 11.014224

Similar Items