Relational Matching with Active Graphs

This paper describes a novel approach to relational matching problems which draws on an active graph paradigm. This active representation is iteratively reconfigured to increase its degree of topological congruency with the model relational structure in a reconstructive matching process. The final restored graph representation is optimal in the sense that it has maximum a posteriori probability with respect to the available attributes for the objects under match. Reassignment of nodes between the graph and an outlier set iteratively optimises the Kullback entropy of the surviving relational cluster. The main benefit of the reconstructive technique over the conventional matching of static relational structures, lies in its rejection of relational noise and an increased robustness to severe levels of scene clutter.