Object Tracking With Structured Metric Learning

被引:0
作者
Zhao, Xiaolin [1 ]
Xu, Zhuofan [2 ]
Zhao, Boxin [1 ]
Chen, Xiaolong [3 ]
Li, Zongzhe [1 ]
机构
[1] Air Force Engn Univ, Equipment Management & UAV Engn Coll, Xian, Shaanxi, Peoples R China
[2] Natl Def Univ, Joint Operat Coll, Shijiazhuang, Hebei, Peoples R China
[3] Flight Automat Control Res Inst, Xian, Shaanxi, Peoples R China
基金
中国国家自然科学基金;
关键词
Target tracking; Learning systems; Training; Deep learning; Robustness; Object tracking; structured metric learning; KNN distance; VISUAL TRACKING; APPEARANCE MODEL;
D O I
10.1109/ACCESS.2019.2950690
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
In this paper, we propose a novel tracking method based on structured metric learning, which takes the advantages of both structured learning and distance metric learning. In our method, tracking is formulated as a structured metric learning problem, which not only considers the importance of different samples, but also improves the discriminability by learning a specific distance metric for matching. Specifically, a concrete structured metric learning method is realized by making use of the constraints from the target and its neighbour training samples under the above framework. Besides, a closed-form solution is derived for the structured metric learning problem. To improve the matching robustness, the K-nearest neighbours (KNN) distance is employed to determine the final tracking result. Experimental results in the benchmark dataset demonstrate that the proposed structured metric learning based tracking method can achieve desirable performance.
引用
收藏
页码:161764 / 161775
页数:12
相关论文
共 50 条
[31]   Robust Object Tracking using Sparse based Representative Dictionary Learning [J].
Jansi, R. ;
Amutha, R. ;
Alice, R. ;
Chitra, E. Mano ;
Ros, G. M. Susmitha .
2017 INTERNATIONAL CONFERENCE ON COMPUTATION OF POWER, ENERGY INFORMATION AND COMMUNICATION (ICCPEIC), 2017, :102-107
[32]   Learning Channel-Aware Correlation Filters for Robust Object Tracking [J].
Nai, Ke ;
Li, Zhiyong ;
Wang, Haidong .
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, 2022, 32 (11) :7843-7857
[33]   Mutual Learning and Feature Fusion Siamese Networks for Visual Object Tracking [J].
Jiang, Min ;
Zhao, Yuyao ;
Kong, Jun .
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, 2021, 31 (08) :3154-3167
[34]   Object Tracking Using Siamese Network-Based Reinforcement Learning [J].
Park, Sung Jun ;
Hwang, Seung Jun ;
Baek, Joong-Hwan .
IEEE ACCESS, 2022, 10 :63339-63352
[35]   Learning Dynamic Spatial-Temporal Regularization for UAV Object Tracking [J].
Deng, Chenwei ;
He, Shuangcheng ;
Han, Yuqi ;
Zhao, Boya .
IEEE SIGNAL PROCESSING LETTERS, 2021, 28 :1230-1234
[36]   Hyperspectral Object Tracking With Context-Aware Learning and Category Consistency [J].
Wang, Ye ;
Mei, Shaohui ;
Ma, Mingyang ;
Liu, Yuheng ;
Gao, Tao ;
Han, Huiyang .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2025, 63
[37]   Robust visual tracking with correlation filters and metric learning [J].
Yuan, Di ;
Kang, Wei ;
He, Zhenyu .
KNOWLEDGE-BASED SYSTEMS, 2020, 195
[38]   Antidecay LSTM for Siamese Tracking With Adversarial Learning [J].
Zhao, Fei ;
Zhang, Ting ;
Wu, Yi ;
Tang, Ming ;
Wang, Jinqiao .
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2021, 32 (10) :4475-4489
[39]   A review of object tracking based on deep learning [J].
Zhao, Guochen ;
Meng, Fanyong ;
Yang, Chengzhuan ;
Wei, Hui ;
Zhang, Dawei ;
Zheng, Zhonglong .
NEUROCOMPUTING, 2025, 651
[40]   COB method with online learning for object tracking [J].
Lin, Luyue ;
Liu, Bo ;
Xiao, Yanshan .
NEUROCOMPUTING, 2020, 393 :142-155