A Hierarchical Bayesian Model for Matching Unlabeled Point Sets

Point set registration is the key in many scientific disciplines. Target at several challenges in registration (e.g. initial registration, outliers, missing data, and local trap), we propose a robust registration method for two point sets using a hierarchical Bayesian model, which is combined with Markov chain Monte Carlo inference. Our approach is based on the introduction of a template of hidden locations underlying the observed configuration points. A Poisson process prior is assigned to these locations, resulting in a simplified formulation of the model. We make use of a structure containing the relevant information on the matches. We conduct several experiments to demonstrate that our algorithm is accurate and robust.