Approximate evaluation of marginal association probabilities with belief propagation

TL;DR

This paper introduces a belief propagation-based method for estimating marginal association probabilities in data association problems, providing convergence guarantees and demonstrating improved accuracy and efficiency over previous approaches.

Contribution

It formulates data association as a graphical model and applies belief propagation with proven convergence and iteration bounds, advancing inference techniques in tracking.

Findings

BP converges with guaranteed bounds

Outperforms prior methods in accuracy

Reduces computational complexity

Abstract

Data association, the problem of reasoning over correspondence between targets and measurements, is a fundamental problem in tracking. This paper presents a graphical model formulation of data association and applies an approximate inference method, belief propagation (BP), to obtain estimates of marginal association probabilities. We prove that BP is guaranteed to converge, and bound the number of iterations necessary. Experiments reveal a favourable comparison to prior methods in terms of accuracy and computational complexity.