DOMOPT: A Detection-Based Online Multi-Object Pedestrian Tracking Network for Videos

被引:1
作者
Huan, Ruohong [1 ]
Zheng, Shuaishuai [1 ]
Xie, Chaojie [1 ]
Chen, Peng [1 ]
Liang, Ronghua [1 ]
机构
[1] Zhejiang Univ Technol, Coll Comp Sci & Technol, Hangzhou, Zhejiang, Peoples R China
基金
中国国家自然科学基金;
关键词
Multi-object tracking; feature fusion; appearance feature; videos; ASSIGNMENT;
D O I
10.1142/S021800142356013X
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Due to the problem of low tracking accuracy and weak tracking stability of current multi-object pedestrian tracking algorithms in complex scenes for videos, a Detection-based Online Multi-Object Pedestrian Tracking (DOMOPT) network is proposed. First, a Multi-Level Feature Fusion (MLFF) pedestrian detection network is proposed based on the Center and Scale Prediction (CSP) algorithm. The pyramid convolutional neural network is used as the backbone to enhance the feature extraction capability for small objects. The shallow features and deep features at multiple levels are integrated to fully obtain the position and semantic information to further improve the detection performance for small objects. Then, on the basis of Joint Detection and Embedding (JDE) architecture, a Multi-Branch Pedestrian Appearance (MBPA) feature extraction network is proposed and added into the pedestrian detection network to extract the appearance feature vector corresponding to each pedestrian. The pedestrian appearance feature extraction is treated as a classification task jointly training with the pedestrian detection task, using the multi-task learning strategy. Experimental results show that the proposed network has better tracking accuracy and stability compared with state-of-the-art algorithms.
引用
收藏
页数:22
相关论文
共 48 条
  • [1] Bergmann P., ARXIV
  • [2] Bewley A, 2016, IEEE IMAGE PROC, P3464, DOI 10.1109/ICIP.2016.7533003
  • [3] Factors Influencing Pediatric Emergency Department Visits for Low-Acuity Conditions
    Long, Christina M.
    Mehrhoff, Casey
    Abdel-Latief, Eman
    Rech, Megan
    Laubham, Matthew
    [J]. PEDIATRIC EMERGENCY CARE, 2021, 37 (05) : 265 - 268
  • [4] Beyond triplet loss: a deep quadruplet network for person re-identification
    Chen, Weihua
    Chen, Xiaotang
    Zhang, Jianguo
    Huang, Kaiqi
    [J]. 30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017), 2017, : 1320 - 1329
  • [5] Learning a Proposal Classifier for Multiple Object Tracking
    Dai, Peng
    Weng, Renliang
    Choi, Wongun
    Zhang, Changshui
    He, Zhangping
    Ding, Wei
    [J]. 2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021, 2021, : 2443 - 2452
  • [6] Pedestrian Detection: An Evaluation of the State of the Art
    Dollar, Piotr
    Wojek, Christian
    Schiele, Bernt
    Perona, Pietro
    [J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2012, 34 (04) : 743 - 761
  • [7] Duta I., ARXIV
  • [8] Ess A, 2008, PROC CVPR IEEE, P1857
  • [9] Learnable Graph Matching: Incorporating Graph Partitioning with Deep Feature Learning for Multiple Object Tracking
    He, Jiawei
    Huang, Zehao
    Wang, Naiyan
    Zhang, Zhaoxiang
    [J]. 2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021, 2021, : 5295 - 5305
  • [10] Hornakova A, 2020, PR MACH LEARN RES, V119