Non-wildcard Matching Beats the Interpretation Tree

Probably the best known control algorithm for high-level model matching in computer vision is the Interpretation Tree expansion algorithm, popularized and extended by Grimson and Lozano-Perez. This algorithm has been shown to have a high computational complexity, particularly when being applied to matching problems with large numbers of features. This paper introduces a non-wildcard variation on this algorithm that has an improvement of about 4-10 in performance over the standard Interpretation Tree algorithm.