1Approximate evaluation of marginal association probabilities with belief propagation
Jason Louis Williams, Roslyn A. Lau · 2016
Abstract—Data association, the problem of reasoning over correspondence between targets and measurements, is a fun-damental problem in tracking tracking. This paper presents a graphical model formulation of data association and applies an approximate inference method, belief propagation (BP), to obtain marginal association probabilities. We prove that BP is guaranteed to converge, and bound the number of iterations required for convergence. Experiments reveal a favourable com-parison to prior methods in terms of accuracy and computational complexity. Index Terms—Data association, tracking, JPDA, graphical models, belief propagation, cycles, convergence guarantees