A New Optimal Subpattern Assignment (OSPA) Metric for Multi-target Filtering

TL;DR

This paper introduces the Complete OSPA (COSPA), a new metric for multi-target filtering that addresses the insensitivity of the original OSPA to empty sets, enabling better performance evaluation.

Contribution

The paper presents COSPA, a novel metric that improves upon OSPA by handling empty sets and separately controlling distance and cardinality errors.

Findings

COSPA retains OSPA's advantages for multi-target filtering evaluation.

COSPA effectively distinguishes between physical distance errors and cardinality errors.

The metric improves performance assessment accuracy in multi-target filtering scenarios.

Abstract

This paper proposes and evaluates a new metric. This metric will overcome a limitation of the Optimal Subpattern Assignment (OSPA) metric mentioned by Schuhmacher et al.: the OSPA distance between two sets of points is insensitive to the the case where one is empty. This proposed metric called Complete OSPA (COSPA), retains all the advantages of the OSPA metric for evaluating the performance of multiple target filtering algorithms while also allowing separate control over the threshold of physical distance errors and cardinality errors.