CSDL Home IEEE Transactions on Pattern Analysis & Machine Intelligence 2009 vol.31 Issue No.02  February
Subscribe
Issue No.02  February (2009 vol.31)
pp: 337350
Adrian M. Peter , University of Florida, Gainesville
Anand Rangarajan , University of Florida Dept. of CISE, Gainesville Gainesville
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/TPAMI.2008.69
ABSTRACT
Shape matching plays a prominent role in the comparison of similar structures. We present a unifying framework for shape matching that uses mixture models to couple both the shape representation and deformation. The theoretical foundation is drawn from information geometry wherein information matrices are used to establish intrinsic distances between parametric densities. When a parameterized probability density function is used to represent a landmarkbased shape, the modes of deformation are automatically established through the information matrix of the density. We first show that given two shapes parameterized by Gaussian mixture models (GMMs), the wellknown Fisher information matrix of the mixture model is also a Riemannian metric (actually, the FisherRao Riemannian metric) and can therefore be used for computing shape geodesics. The FisherRao metric has the advantage of being an intrinsic metric and invariant to reparameterization. The geodesic?computed using this metric?establishes an intrinsic deformation between the shapes, thus unifying both shape representation and deformation. A fundamental drawback of the FisherRao metric is that it is not available in closed form for the GMM. Consequently, shape comparisons are computationally very expensive. To address this, we develop a new Riemannian metric based on generalized \phientropy measures. In sharp contrast to the FisherRao metric, the new metric is available in closed form. Geodesic computations using the new metric are considerably more efficient. We validate the performance and discriminative capabilities of these new information geometrybased metrics by pairwise matching of corpus callosum shapes. We also study the deformations of fish shapes that have various topological properties. A comprehensive comparative analysis is also provided using other landmarkbased distances, including the Hausdorff distance, the Procrustes metric, landmarkbased diffeomorphisms, and the bending energies of the thinplate (TPS) and Wendland splines.
INDEX TERMS
Information geometry, Fisher information, FisherRao metric, HavrdaCharvá t entropy, Gaussian mixture models, shape analysis, shape matching, landmark shapes.
CITATION
Adrian M. Peter, Anand Rangarajan, "Information Geometry for Landmark Shape Analysis: Unifying Shape Representation and Deformation", IEEE Transactions on Pattern Analysis & Machine Intelligence, vol.31, no. 2, pp. 337350, February 2009, doi:10.1109/TPAMI.2008.69
REFERENCES
