Adaptive multi-object tracking based on sensors fusion with confidence updating

• Published in: International journal of applied earth observation and geoinformation Vol. 125; p. 103577
• Main Authors: Liu, Junting, Liu, Deer, Ji, Weizhen, Cai, Chengfeng, Liu, Zhen
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.12.2023; Elsevier
• Subjects: 3D multi-object tracking; Adaptive factor; Confidence; Sensors fusion; Sensors fusion; Confidence; 3D multi-object tracking; Adaptive factor
• ISSN: 1569-8432; 1872-826X
• DOI: 10.1016/j.jag.2023.103577

•We have improved the Kalman filter and data association algorithm based on detection confidence, reducing the impact of dynamic errors.•A simple and effective fusion method of LiDAR and camera data is proposed. This fusion addresses the issue of a large number of false detections commonly encountered in SOTA methods.•We discussed the methods based on nonlinear motion models and improve them to solve the problem of trajectory deviation during intelligent vehicle turning. Multi-object tracking (MOT) systems typically rely on object detection results for tracking, so the accuracy of the MOT system is significantly affected by the error of the detector. Changes in error usually lead to unstable tracking. Regarding this problem, we proposed an adaptive MOT method based on detection confidence. At first, we use a simple data fusion method to combine the detection results of LiDAR and camera to reduce the large number of false detections. And then we used a factor based on confidence to adjust the estimating covariance matrix and measurement covariance matrix adaptively. The algorithm can judge which is more reliable between prediction and detection, and choose which is more important in the update step. Meanwhile, we set a factor based on confidence to control the search range in the data association module. Our method reduces the impact of detector error while ensuring accuracy and speed, and improves the robustness of the MOT algorithm. Through experiments conducted on the KITTI multi-object tracking dataset, our method has demonstrated significant advantages over state-of-the-art (SOTA) methods in terms of both accuracy and processing speed. The results of MOTA for 90.02% and FPS for 262.