Fusing target information from multiple views for robust visual tracking
In this study, the authors address the problem of tracking a single target in a calibrated multi‐camera surveillance system with information on its location in the first frame of each view. Recently, tracking with online multiple instance learning (OMIL) has been shown to give promising tracking res...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Wiley
2014-04-01
|
Series: | IET Computer Vision |
Subjects: | |
Online Access: | https://doi.org/10.1049/iet-cvi.2013.0026 |
_version_ | 1797685253285871616 |
---|---|
author | Keli Hu Xing Zhang Yuzhang Gu Yingguan Wang |
author_facet | Keli Hu Xing Zhang Yuzhang Gu Yingguan Wang |
author_sort | Keli Hu |
collection | DOAJ |
description | In this study, the authors address the problem of tracking a single target in a calibrated multi‐camera surveillance system with information on its location in the first frame of each view. Recently, tracking with online multiple instance learning (OMIL) has been shown to give promising tracking results. However, it may fail in a real surveillance system because of problems arising from target orientation, scale or illumination changes. In this study, the authors show that fusing target information from multiple views can avoid these problems and lead to a more robust tracker. At each camera node, an efficient OMIL algorithm is used to model target appearance. To update the OMIL‐based classifier in one view, a co‐training strategy is applied to generate a representative set of training bags from all views. Bags extracted from each view hold a unique weight depending on similarity of target appearance between the current view and the view which contains the classifier that needs to be updated. In addition, target motion on a camera's image plane is modelled by a modified particle filter guided by the corresponding object two‐dimensional (2D) location and fused 3D location. Experimental results demonstrate that the proposed algorithm is robust for human tracking in challenging scenes. |
first_indexed | 2024-03-12T00:41:29Z |
format | Article |
id | doaj.art-aee3defd175f4547a558aa10ec3d69b7 |
institution | Directory Open Access Journal |
issn | 1751-9632 1751-9640 |
language | English |
last_indexed | 2024-03-12T00:41:29Z |
publishDate | 2014-04-01 |
publisher | Wiley |
record_format | Article |
series | IET Computer Vision |
spelling | doaj.art-aee3defd175f4547a558aa10ec3d69b72023-09-15T07:13:10ZengWileyIET Computer Vision1751-96321751-96402014-04-0182869710.1049/iet-cvi.2013.0026Fusing target information from multiple views for robust visual trackingKeli Hu0Xing Zhang1Yuzhang Gu2Yingguan Wang3Key Laboratory of Wireless Sensor Network & CommunicationShanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences865 Changning RoadChangning DistrictShanghai200050People's Republic of ChinaKey Laboratory of Wireless Sensor Network & CommunicationShanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences865 Changning RoadChangning DistrictShanghai200050People's Republic of ChinaKey Laboratory of Wireless Sensor Network & CommunicationShanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences865 Changning RoadChangning DistrictShanghai200050People's Republic of ChinaKey Laboratory of Wireless Sensor Network & CommunicationShanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences865 Changning RoadChangning DistrictShanghai200050People's Republic of ChinaIn this study, the authors address the problem of tracking a single target in a calibrated multi‐camera surveillance system with information on its location in the first frame of each view. Recently, tracking with online multiple instance learning (OMIL) has been shown to give promising tracking results. However, it may fail in a real surveillance system because of problems arising from target orientation, scale or illumination changes. In this study, the authors show that fusing target information from multiple views can avoid these problems and lead to a more robust tracker. At each camera node, an efficient OMIL algorithm is used to model target appearance. To update the OMIL‐based classifier in one view, a co‐training strategy is applied to generate a representative set of training bags from all views. Bags extracted from each view hold a unique weight depending on similarity of target appearance between the current view and the view which contains the classifier that needs to be updated. In addition, target motion on a camera's image plane is modelled by a modified particle filter guided by the corresponding object two‐dimensional (2D) location and fused 3D location. Experimental results demonstrate that the proposed algorithm is robust for human tracking in challenging scenes.https://doi.org/10.1049/iet-cvi.2013.0026human trackingfused 3D location2D locationobject two-dimensional locationparticle filtertarget motion |
spellingShingle | Keli Hu Xing Zhang Yuzhang Gu Yingguan Wang Fusing target information from multiple views for robust visual tracking IET Computer Vision human tracking fused 3D location 2D location object two-dimensional location particle filter target motion |
title | Fusing target information from multiple views for robust visual tracking |
title_full | Fusing target information from multiple views for robust visual tracking |
title_fullStr | Fusing target information from multiple views for robust visual tracking |
title_full_unstemmed | Fusing target information from multiple views for robust visual tracking |
title_short | Fusing target information from multiple views for robust visual tracking |
title_sort | fusing target information from multiple views for robust visual tracking |
topic | human tracking fused 3D location 2D location object two-dimensional location particle filter target motion |
url | https://doi.org/10.1049/iet-cvi.2013.0026 |
work_keys_str_mv | AT kelihu fusingtargetinformationfrommultipleviewsforrobustvisualtracking AT xingzhang fusingtargetinformationfrommultipleviewsforrobustvisualtracking AT yuzhanggu fusingtargetinformationfrommultipleviewsforrobustvisualtracking AT yingguanwang fusingtargetinformationfrommultipleviewsforrobustvisualtracking |