A Metric on the Space of Finite Sets of Trajectories for Evaluation of Multi-Target Tracking Algorithms

• Published in: IEEE transactions on signal processing Vol. 68; pp. 3917 - 3928
• Main Authors: Garcia-Fernandez, Angel F., Rahmathullah, Abu Sajana, Svensson, Lennart
• Format: Journal Article
• Language: English
• Published: New York IEEE 2020; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Error detection; Ground truth; Linear programming; Lower bounds; Metrics; Monte Carlo methods; Multiple target tracking; Operations research; Polynomials; random finite sets; sets of trajectories; Signal processing algorithms; Switches; Target detection; Target tracking; Trajectory; Trajectory analysis
• ISSN: 1053-587X; 1941-0476; 1941-0476
• DOI: 10.1109/TSP.2020.3005309

In this article, we propose a metric on the space of finite sets of trajectories for assessing multi-target tracking algorithms in a mathematically sound way. The main use of the metric is to compare estimates of trajectories from different algorithms with the ground truth of trajectories. The proposed metric includes intuitive costs associated to localization error for properly detected targets, missed and false targets and track switches at each time step. The metric computation is based on solving a multi-dimensional assignment problem. We also propose a lower bound for the metric, which is also a metric for sets of trajectories and is computable in polynomial time using linear programming. We also extend the proposed metrics on sets of trajectories to random finite sets of trajectories.