Performance evaluation of a multiple-hypothesis multi-target tracking algorithm

Author(s):  
K.-C. Chang ◽  
S. Mori ◽  
C.-Y. Chong
2013 ◽  
Vol 694-697 ◽  
pp. 2341-2344
Author(s):  
Shu Rong Tian ◽  
Xiao Shu Sun ◽  
Dan Liu

This paper is concerned with the performance evaluation of algorithm of multi-target and target types tracking. Performance evaluation is based on information theory, Kullback-Leibler measure is used to discriminate information provided by algorithm. Through simulations, algorithm of multi-target tracking was evaluated in term of information (localization, classification, and target number components) the algorithm provide about the actual state of ground truth.


2021 ◽  
Author(s):  
Tingting Kou ◽  
Hua Cai ◽  
Guangwen Liu ◽  
Yingchao Li

Author(s):  
Andinet Hunde ◽  
Beshah Ayalew

Target tracking in public traffic calls for a tracking system with automated track initiation and termination facilities in a randomly evolving driving environment. In addition, the key problem of data association needs to be handled effectively considering the limitations in the computational resources onboard an autonomous car. In this paper, we discuss a multi-target tracking system that addresses target birth/initiation and death/termination processes with automatic track management feature. The tracking system is based on Linear Multi-target Integrated Probabilistic Data Association Filter (LMIPDAF), which is adapted to specifically include algorithms that handle track initiation and termination, clutter density estimation and track management. The performance of the proposed tracking algorithm is compared to other single and multi-target tracking schemes and is shown to have acceptable tracking error. It is further illustrated through multiple traffic simulations that the computational requirement of the tracking algorithm is less than that of optimal multi-target tracking algorithms that explicitly address data association uncertainties.


Sign in / Sign up

Export Citation Format

Share Document