Learning Spatio-Temporal Information for Multi-Object Tracking
The robust multi-object tracking problem is a challenging issue in the field of computer vision. In this paper, we propose a multi-object tracking algorithm with temporal-spatial information and trajectory of confidence. The whole process is divided into local and global association. Trajectories wi...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2017-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/7885605/ |
id |
doaj-c9fd9230616547769759ba8157544250 |
---|---|
record_format |
Article |
spelling |
doaj-c9fd9230616547769759ba81575442502021-03-29T20:11:13ZengIEEEIEEE Access2169-35362017-01-0153869387710.1109/ACCESS.2017.26864827885605Learning Spatio-Temporal Information for Multi-Object TrackingJian Wei0https://orcid.org/0000-0001-8952-7703Mei Yang1Feng Liu2Jiangsu Province Key Laboratory on Image Processing and Image Communications, Nanjing University of Posts and Telecommunications, Nanjing, ChinaJiangsu Province Key Laboratory on Image Processing and Image Communications, Nanjing University of Posts and Telecommunications, Nanjing, ChinaJiangsu Province Key Laboratory on Image Processing and Image Communications, Nanjing University of Posts and Telecommunications, Nanjing, ChinaThe robust multi-object tracking problem is a challenging issue in the field of computer vision. In this paper, we propose a multi-object tracking algorithm with temporal-spatial information and trajectory of confidence. The whole process is divided into local and global association. Trajectories with high confidence are associated with the detection result of the current frame during local association, whereas trajectories with low confidence are associated with the detection results of the current frame are not matched during global association. We determine the association results using a combined model. By utilizing the information of spatial-temporal correlation, the model is more robust and can deal with missed detection. In addition, we measure the reliability of the spatial information by the confidence map smoothing constraint and the peak sidelobe ratio criterion. We conduct experiments using a challenging public data set, and the results show that our proposed algorithm is superior to many other popular algorithms when dealing with problems, such as missed detection and poor tracker robustness.https://ieeexplore.ieee.org/document/7885605/Multi-object trackingtrajectory of confidencespatio-temporal information |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Jian Wei Mei Yang Feng Liu |
spellingShingle |
Jian Wei Mei Yang Feng Liu Learning Spatio-Temporal Information for Multi-Object Tracking IEEE Access Multi-object tracking trajectory of confidence spatio-temporal information |
author_facet |
Jian Wei Mei Yang Feng Liu |
author_sort |
Jian Wei |
title |
Learning Spatio-Temporal Information for Multi-Object Tracking |
title_short |
Learning Spatio-Temporal Information for Multi-Object Tracking |
title_full |
Learning Spatio-Temporal Information for Multi-Object Tracking |
title_fullStr |
Learning Spatio-Temporal Information for Multi-Object Tracking |
title_full_unstemmed |
Learning Spatio-Temporal Information for Multi-Object Tracking |
title_sort |
learning spatio-temporal information for multi-object tracking |
publisher |
IEEE |
series |
IEEE Access |
issn |
2169-3536 |
publishDate |
2017-01-01 |
description |
The robust multi-object tracking problem is a challenging issue in the field of computer vision. In this paper, we propose a multi-object tracking algorithm with temporal-spatial information and trajectory of confidence. The whole process is divided into local and global association. Trajectories with high confidence are associated with the detection result of the current frame during local association, whereas trajectories with low confidence are associated with the detection results of the current frame are not matched during global association. We determine the association results using a combined model. By utilizing the information of spatial-temporal correlation, the model is more robust and can deal with missed detection. In addition, we measure the reliability of the spatial information by the confidence map smoothing constraint and the peak sidelobe ratio criterion. We conduct experiments using a challenging public data set, and the results show that our proposed algorithm is superior to many other popular algorithms when dealing with problems, such as missed detection and poor tracker robustness. |
topic |
Multi-object tracking trajectory of confidence spatio-temporal information |
url |
https://ieeexplore.ieee.org/document/7885605/ |
work_keys_str_mv |
AT jianwei learningspatiotemporalinformationformultiobjecttracking AT meiyang learningspatiotemporalinformationformultiobjecttracking AT fengliu learningspatiotemporalinformationformultiobjecttracking |
_version_ |
1724195136863207424 |