Articulated shape matching using Laplacian eigenfunctions and unsupervised point registration

Matching articulated shapes represented by voxel-sets reduces to maximal sub-graph isomorphism when each set is described by a weighted graph. Spectral graph theory can be used to map these graphs onto lower dimensional spaces and match shapes by aligning their embeddings in virtue of their invariance to change of pose. Classical graph isomorphism schemes relying on the ordering of the eigenvalues to align the eigenspaces fail when handling large data-sets or noisy data. We derive a new formulation that finds the best alignment between two congruent K-dimensional sets of points by selecting the best subset of eigenfunctions of the Laplacian matrix. The selection is done by matching eigenfunction signatures built with histograms, and the retained set provides a smart initialization for the alignment problem with a considerable impact on the overall performance. Dense shape matching casted into graph matching reduces then, to point registration of embeddings under orthogonal transformations; the registration is solved using the framework of unsupervised clustering and the EM algorithm. Maximal subset matching of non identical shapes is handled by defining an appropriate outlier class. Experimental results on challenging examples show how the algorithm naturally treats changes of topology, shape variations and different sampling densities.

[1]  K. S. Arun,et al.  Least-Squares Fitting of Two 3-D Point Sets , 1987, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2]  Shinji Umeyama,et al.  An Eigendecomposition Approach to Weighted Graph Matching Problems , 1988, IEEE Trans. Pattern Anal. Mach. Intell..

[3]  H. C. Longuet-Higgins,et al.  An algorithm for associating the features of two images , 1991, Proceedings of the Royal Society of London. Series B: Biological Sciences.

[4]  Michael Brady,et al.  Feature-based correspondence: an eigenvector approach , 1992, Image Vis. Comput..

[5]  G. McLachlan,et al.  The EM algorithm and extensions , 1996 .

[6]  S T Roweis,et al.  Nonlinear dimensionality reduction by locally linear embedding. , 2000, Science.

[7]  Michael I. Jordan,et al.  On Spectral Clustering: Analysis and an algorithm , 2001, NIPS.

[8]  Edwin R. Hancock,et al.  Spectral correspondence for point pattern matching , 2003, Pattern Recognit..

[9]  Edwin R. Hancock,et al.  A unified framework for alignment and correspondence , 2003, Comput. Vis. Image Underst..

[10]  Edwin R. Hancock,et al.  Correspondence Matching with Modal Clusters , 2003, IEEE Trans. Pattern Anal. Mach. Intell..

[11]  Mikhail Belkin,et al.  Laplacian Eigenmaps for Dimensionality Reduction and Data Representation , 2003, Neural Computation.

[12]  Maja J. Mataric,et al.  Markerless kinematic model and motion capture from volume sequences , 2003, 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2003. Proceedings..

[13]  Ron Kimmel,et al.  On Bending Invariant Signatures for Surfaces , 2003, IEEE Trans. Pattern Anal. Mach. Intell..

[14]  Sebastian Thrun,et al.  The Correlated Correspondence Algorithm for Unsupervised Registration of Nonrigid Surfaces , 2004, NIPS.

[15]  William M. Wells,et al.  Statistical Approaches to Feature-Based Object Recognition , 2004, International Journal of Computer Vision.

[16]  Zhengyou Zhang,et al.  Iterative point matching for registration of free-form curves and surfaces , 1994, International Journal of Computer Vision.

[17]  Hao Zhang,et al.  Robust 3D Shape Correspondence in the Spectral Domain , 2006, IEEE International Conference on Shape Modeling and Applications 2006 (SMI'06).

[18]  Radu Horaud,et al.  The Alignment Between 3-D Data and Articulated Shapes with Bending Surfaces , 2006, ECCV.

[19]  Radford M. Neal Pattern Recognition and Machine Learning , 2007, Technometrics.

[20]  Adrian Hilton,et al.  Correspondence labelling for wide-timeframe free-form surface matching , 2007, 2007 IEEE 11th International Conference on Computer Vision.