Sparse point cloud multi-target tracking method fusing spatio-temporal information

Abstract

The invention provides a sparse point cloud multi-target tracking method fusing spatio-temporal information, and belongs to the field of 3D vision. According to the method, the point cloud feature extraction network is used as a trunk, multiple frames of point cloud data are input at the same time, time domain information fusion is carried out on extracted features, and therefore missing detectioncaused by point cloud sparseness is avoided. Benefited from the fusion of time-space information, tracking and detection tasks can be more tightly coupled, and detection frames of three frames beforeand after can be predicted at the same time to obtain a track segment of a current target for three continuous frames. Then a distance intersection ratio score of the current trajectory segment and the trajectory tracking result at the previous moment is calculated, and the currently split trajectory segment and the historical trajectory segment are matched and spliced by using a greedy algorithmto obtain a final trajectory tracking result at each moment. The method has the application potential of coping with multi-target tracking under sparse point clouds, has high robustness for target missing detection and false detection, and can still obtain a stable tracking result in sparse point cloud sequence input.

Description

technical field [0001] The invention relates to the field of 3D vision, in particular to a sparse point cloud multi-target tracking method for fusing spatio-temporal information. Background technique [0002] In recent years, the rapid development of autonomous driving, robotics and other fields has put forward an urgent need for high-precision target detection and tracking algorithms, so 3D target detection and tracking is becoming the most noteworthy research direction in computer vision. The detection and tracking task takes the raw data of the sensor as input, and outputs the accurate target position and tracking id, which is the basis of subsequent links such as path planning and an essential part of the entire system. Faced with the demand for precise three-dimensional positioning and tracking, depth cameras or multi-camera sensors have low precision, short positioning distances, and are greatly affected by light. LiDAR has the characteristics of long distance, high p...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to propose a sparse point cloud multi-target tracking method that integrates spatio-temporal information in order to overcome the problems of low accuracy and insufficient information utilization of the existing 3D multi-target tracking algorithm under sparse point cloud data

Images