Unmanned multi target tracking based on boundary IoU target association
Abstract
Aiming at the defects of target matching measurement and fixed life cycle management in traditional multi-target tracking, a target association and adaptive life cycle management strategy based on boundary IoU is proposed. Boundary IoU takes into account the advantages of Euclidean distance and IoU, which can improve the accuracy of target matching. The adaptive life cycle management correlates the target trajectory confidence with the life cycle, which significantly reduces the target loss and trajectory false detection during the tracking process. Experiments on KITTI multi-target tracking dataset demonstrate the effectiveness of the proposed method.
References
[1] Liu Yuchen, Tang Xing, Su Yan. Real-time target 3D detection method of unmanned platform based on camera and LiDAR
:CN201911244310.7[P].CN110879401A[2023-11-02].
[2] Cheng Xiao-qiang. Unmanned 3D target detection and multi-target tracking based on LiDAR [D]. Tianjin University,2020.
[3] Cao Weihao. Ground Target Detection method and Application based on 3D LiDAR [J].[2023-11-02].
[4] Liu Mingyu, Yan Jun, GE Wancheng. Review of 3D target detection technology based on LiDAR point cloud [J]. Television Technology,2022(001):046.
[5] Zhang Pu, LIU JQ, XIAO Jchao, et al. Target location and tracking based on fusion of camera and LiDAR [J]. Advances in Laser
and Optoelectronics, 2019,61(08).
[6] Zhang Xiangli. Research on surface vessel detection and recognition based on LiDAR and visual information fusion [D]. Shanghai
Jiao Tong University,2017.
[7] Xiong Zhenkai, Cheng Xiaoqiang, Wu Youdong, et al. Based on unmanned 3d laser radar multi-target tracking [J]. Journal of automation, 2022,: 1-11. DOI: 10.16383 / j.a as c210783.
[8] Jiang Jingwen, Zhu Bing, Ma Tianfei, et al. A 3D target detection and tracking method based on camera and LiDAR
:CN202011337882.2[P].CN112487919A[2023-11-02].
[9] Wen Long. Research on obstacle detection and tracking technology of unmanned vehicle based on vehicle Lidar [D]. Jilin Universi_x005fty,2020.
[10] Wang Qing-Lin, Li Hui, Xie Li-Zhi, et al. Research on Improvement of vehicle target detection Algorithm based on LiDAR point
Cloud [J]. Electronic Measurement Technology,2023,46(1):120-126.
