Research on Multi-Objective Pedestrian Tracking Algorithm Based on Full-Size Feature Fusion

In the field of Multi-Object Tracking (MOT), the current mainstream approach is the tracking by detection paradigm, which heavily relies on the accuracy of the detector, the comprehensiveness of feature extraction, and the superiority of the data association matching algorithm. Most existing pedestr...

Full description

Saved in:
Bibliographic Details
Main Authors: Yihuai Zhu, Zhandong Liu, Ke Li, Yong Li, Xiangwei Qi, Nan Ding
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10908239/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849772690592759808
author Yihuai Zhu
Zhandong Liu
Ke Li
Yong Li
Xiangwei Qi
Nan Ding
author_facet Yihuai Zhu
Zhandong Liu
Ke Li
Yong Li
Xiangwei Qi
Nan Ding
author_sort Yihuai Zhu
collection DOAJ
description In the field of Multi-Object Tracking (MOT), the current mainstream approach is the tracking by detection paradigm, which heavily relies on the accuracy of the detector, the comprehensiveness of feature extraction, and the superiority of the data association matching algorithm. Most existing pedestrian re-identification methods are based on convolutional neural networks (CNNs), which struggle to balance both local and global features of pedestrians. Given that current detectors are already highly advanced, this paper proposes a full-scale feature fusion-based multi-object pedestrian tracking algorithm named BOS-SORT. The algorithm utilizes the proposed feature extraction network, Better Omni-Scale Net (BOSNet), to captures both global and local appearance information, effectively reducing appearance information loss. Furthermore, it employs an improved association matching algorithm, AveSort, to combines IoU and appearance features for initial data association while smoothing target motion states to minimize matching errors in high-similarity scenarios. The BOS-SORT system integrates these methods and demonstrates exceptional capability in aligning global trajectories with real trajectories. Experimental results show that it achieves state-of-the-art Higher Order Tracking Accuracy (HOTA) scores of 66.2 and 65.3 on the MOT17 and MOT20 datasets, respectively.
format Article
id doaj-art-c72fd93dbc224a12a9de504c07feb7a0
institution DOAJ
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-c72fd93dbc224a12a9de504c07feb7a02025-08-20T03:02:15ZengIEEEIEEE Access2169-35362025-01-0113391073911610.1109/ACCESS.2025.354681810908239Research on Multi-Objective Pedestrian Tracking Algorithm Based on Full-Size Feature FusionYihuai Zhu0https://orcid.org/0009-0002-8728-7069Zhandong Liu1https://orcid.org/0000-0002-6073-4618Ke Li2Yong Li3Xiangwei Qi4Nan Ding5School of Computer Science and Technology, Xinjiang Normal University, Ürümqi, ChinaSchool of Computer Science and Technology, Xinjiang Normal University, Ürümqi, ChinaSchool of Computer Science and Technology, Xinjiang Normal University, Ürümqi, ChinaSchool of Computer Science and Technology, Xinjiang Normal University, Ürümqi, ChinaSchool of Computer Science and Technology, Xinjiang Normal University, Ürümqi, ChinaSchool of Computer Science and Technology, Xinjiang Normal University, Ürümqi, ChinaIn the field of Multi-Object Tracking (MOT), the current mainstream approach is the tracking by detection paradigm, which heavily relies on the accuracy of the detector, the comprehensiveness of feature extraction, and the superiority of the data association matching algorithm. Most existing pedestrian re-identification methods are based on convolutional neural networks (CNNs), which struggle to balance both local and global features of pedestrians. Given that current detectors are already highly advanced, this paper proposes a full-scale feature fusion-based multi-object pedestrian tracking algorithm named BOS-SORT. The algorithm utilizes the proposed feature extraction network, Better Omni-Scale Net (BOSNet), to captures both global and local appearance information, effectively reducing appearance information loss. Furthermore, it employs an improved association matching algorithm, AveSort, to combines IoU and appearance features for initial data association while smoothing target motion states to minimize matching errors in high-similarity scenarios. The BOS-SORT system integrates these methods and demonstrates exceptional capability in aligning global trajectories with real trajectories. Experimental results show that it achieves state-of-the-art Higher Order Tracking Accuracy (HOTA) scores of 66.2 and 65.3 on the MOT17 and MOT20 datasets, respectively.https://ieeexplore.ieee.org/document/10908239/Multi-object trackingobject detectionperson re-identificationpedestrian trackingdata association
spellingShingle Yihuai Zhu
Zhandong Liu
Ke Li
Yong Li
Xiangwei Qi
Nan Ding
Research on Multi-Objective Pedestrian Tracking Algorithm Based on Full-Size Feature Fusion
IEEE Access
Multi-object tracking
object detection
person re-identification
pedestrian tracking
data association
title Research on Multi-Objective Pedestrian Tracking Algorithm Based on Full-Size Feature Fusion
title_full Research on Multi-Objective Pedestrian Tracking Algorithm Based on Full-Size Feature Fusion
title_fullStr Research on Multi-Objective Pedestrian Tracking Algorithm Based on Full-Size Feature Fusion
title_full_unstemmed Research on Multi-Objective Pedestrian Tracking Algorithm Based on Full-Size Feature Fusion
title_short Research on Multi-Objective Pedestrian Tracking Algorithm Based on Full-Size Feature Fusion
title_sort research on multi objective pedestrian tracking algorithm based on full size feature fusion
topic Multi-object tracking
object detection
person re-identification
pedestrian tracking
data association
url https://ieeexplore.ieee.org/document/10908239/
work_keys_str_mv AT yihuaizhu researchonmultiobjectivepedestriantrackingalgorithmbasedonfullsizefeaturefusion
AT zhandongliu researchonmultiobjectivepedestriantrackingalgorithmbasedonfullsizefeaturefusion
AT keli researchonmultiobjectivepedestriantrackingalgorithmbasedonfullsizefeaturefusion
AT yongli researchonmultiobjectivepedestriantrackingalgorithmbasedonfullsizefeaturefusion
AT xiangweiqi researchonmultiobjectivepedestriantrackingalgorithmbasedonfullsizefeaturefusion
AT nanding researchonmultiobjectivepedestriantrackingalgorithmbasedonfullsizefeaturefusion