US 12,243,242 B2
Multi-target tracking method, device and computer-readable storage medium
Shuping Hu, Shenzhen (CN); Jun Cheng, Shenzhen (CN); Jingtao Zhang, Shenzhen (CN); Miaochen Guo, Shenzhen (CN); Dong Wang, Shenzhen (CN); Zaiwang Gu, Shenzhen (CN); and Jianxin Pang, Shenzhen (CN)
Assigned to UBTECH ROBOTICS CORP LTD, Shenzhen (CN)
Filed by UBTECH ROBOTICS CORP LTD, Shenzhen (CN)
Filed on Jul. 18, 2022, as Appl. No. 17/866,574.
Application 17/866,574 is a continuation of application No. PCT/CN2021/131679, filed on Nov. 19, 2021.
Claims priority of application No. 202110343336.8 (CN), filed on Mar. 30, 2021.
Prior Publication US 2022/0375106 A1, Nov. 24, 2022
Int. Cl. G06T 7/246 (2017.01); G06T 7/70 (2017.01); G06T 7/73 (2017.01)
CPC G06T 7/246 (2017.01) [G06T 7/73 (2017.01); G06T 7/70 (2017.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented multi-target tracking method, comprising:
performing target detection on a current image to obtain detection information of a plurality of detected targets, wherein the detection information comprises position detection information of the detected targets;
obtaining position prediction information of each of a plurality of tracked targets and a number of times of tracking losses of targets from tracking information of each of the tracked targets, and determining a first matching threshold for each of the tracked targets according to the number of times of tracking losses of targets;
calculating a motion matching degree between each of the tracked targets and each of the detected targets according to the position detection information and the position prediction information;
for each of the tracked targets, obtaining a motion matching result according to the motion matching degree and the first matching threshold corresponding to the tracked target; and
matching the detected targets and the tracked targets according to the motion matching results to obtain a tracking result;
wherein the detection information further comprises first appearance features of the detected targets, and the tracking information comprises second appearance features of the tracked targets, matching the detected targets and the tracked targets according to the motion matching result to obtain the tracking result, comprises:
calculating a maximum appearance matching degree between each of the tracked targets and each of the detected targets according to the first appearance features and the second appearance features;
for each of the tracked targets, obtaining an appearance matching result according to the maximum appearance matching degree and a second matching threshold corresponding to the tracked target; and
matching the detected targets and the tracked targets according to the motion matching results and the appearance matching results to obtain the tracking result; and
wherein matching the detected targets and the tracked targets according to the motion matching results to obtain the tracking result, comprises:
for one of the detected targets and one of the tracked targets that are matched in motion information and appearance information thereof, calculating a total matching degree between the one of the detected targets and the one of the tracked targets according to the motion matching degree and the maximum appearance matching degree between the one of the detected targets and the one of the tracked targets; and
performing a matching based on the total matching degree, and determining matching tracked targets and detected targets, using a cascade matching mechanism and a minimum cost algorithm.