A new performance metric for multiple target tracking based on optimal subpattern assignment

This paper proposes a new metric based on the Optimal Sub-pattern Assignment (OSPA) metric for evaluating the performance of multiple target tracking algorithms. It is shown in this paper that by considering the properties of false tracks, missing tracks and miss-detections, the minimization of all distances between tracks enables performance evaluation of multi-target algorithms in a more comprehensive manner than is possible with the OSPA metric for track (OSPAT) introduced by Ristic et al.