Variational Probabilistic Multi-Hypothesis Tracking

TL;DR

This paper introduces VPMHT, a variational Bayesian approach to multi-target tracking that improves track-loss handling and accuracy over traditional methods, validated through extensive simulations.

Contribution

The paper presents a novel VPMHT algorithm based on VBEM, enhancing multi-target tracking by better managing track-loss with variational inference.

Findings

VPMHT outperforms traditional PMHT in track-loss scenarios

VPMHT achieves comparable or better tracking accuracy

Numerical simulations confirm the effectiveness of VPMHT

Abstract

This paper proposes a novel multi-target tracking (MTT) algorithm for scenarios with arbitrary numbers of measurements per target. We propose the variational probabilistic multi-hypothesis tracking (VPMHT) algorithm based on the variational Bayesian expectation-maximisation (VBEM) algorithm to resolve the MTT problem in the classic PMHT algorithm. With the introduction of variational inference, the proposed VPMHT handles track-loss much better than the conventional probabilistic multi-hypothesis tracking (PMHT) while preserving a similar or even better tracking accuracy. Extensive numerical simulations are conducted to demonstrate the effectiveness of the proposed algorithm.