最优 OSPA 估计中的幽灵效应及其由 GOSPA 的解决
信号处理
2019-08-26 v1 计算机视觉与模式识别
机器学习
摘要
本文中,我们展示了在基于最优子模式分配(OSPA)度量的多目标最优估计中所出现的“远距离幽灵效应”。该效应指:若我们在相距遥远的位置有多个独立的潜在目标,其中某一个存在概率的改变会彻底改变其余潜在目标的最优估计。与 OSPA 不同,广义 OSPA(GOSPA)度量()对正确检测目标、虚假目标与漏检目标的定位误差进行惩罚。因此,最优 GOSPA 估计旨在减少虚假与漏检目标的数量,以及正确检测目标的定位误差,并避免幽灵效应。
引用
@article{arxiv.1908.08815,
title = {Spooky effect in optimal OSPA estimation and how GOSPA solves it},
author = {Ángel F. García-Fernández and Lennart Svensson},
journal= {arXiv preprint arXiv:1908.08815},
year = {2019}
}
备注
This paper received the third best paper award at the 22nd International Conference on Information Fusion, Ottawa, Canada, 2019. Matlab code of the GOSPA metric can be found in https://github.com/abusajana/GOSPA . Additional information on MTT can be found in the online course https://www.youtube.com/channel/UCa2-fpj6AV8T6JK1uTRuFpw