Unsupervised Clustering of Shapes

A new method for unsupervised clustering of shapes is here proposed. This method is based on two steps: in the first step a preliminary clusterization is obtained by considering the distance among shapes after alignment with procrustes analysis [1],[2]. This step is based on the minimization of the functional θ(N cluster ) =αN cluster + (1/N cluster )dist(c i ) where N cluster is the total number of clusters, dist(c i ) is the intra-cluster variability and α is an appropriate constant. In the second step, the curvature of shapes belonging to clusters obtained in the first step is examined to i) identify possible outliers and to ii) introduce a further refinement of clusters. The proposed method was tested on the Kimia, Surrey and MPEG7 shape databases and was able to obtain correct clusters, corresponding to perceptually homogeneous object categories. The proposed method was able to distinguish shapes with subtle differences, such as birds with one or two feet and to distinguish among very similar animal species....

[1]  Wesley E. Snyder,et al.  Application of Affine-Invariant Fourier Descriptors to Recognition of 3-D Objects , 1990, IEEE Trans. Pattern Anal. Mach. Intell..

[2]  Hiroshi Murase,et al.  Visual learning and recognition of 3-d objects from appearance , 2005, International Journal of Computer Vision.

[3]  G. Medioni,et al.  Corner detection and curve representation using cubic B-splines , 1986, Proceedings. 1986 IEEE International Conference on Robotics and Automation.

[4]  C. Goodall.  Procrustes methods in the statistical analysis of shape , 1991 .

[5]  F. Bookstein.  Landmark methods for forms without landmarks: localizing group differences in outline shape , 1996, Proceedings of the Workshop on Mathematical Methods in Biomedical Image Analysis.

[6]  Josef Kittler,et al.  Efficient and Robust Retrieval by Shape Content through Curvature Scale Space , 1998, Image Databases and Multi-Media Search.

[7]  Anil K. Jain,et al.  Data clustering: a review , 1999, CSUR.

[8]  Timothy F. Cootes,et al.  Active Shape Models-Their Training and Application , 1995, Comput. Vis. Image Underst..

[9]  Shimon Edelman,et al.  Representation and recognition in vision , 1999 .

[10]  Anuj Srivastava,et al.  Statistical shape analysis: clustering, learning, and testing , 2005, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[11]  James M. Coggins.  Statistical approach to multiscale medial vision , 1992, Optics & Photonics.

[12]  Philip N. Klein,et al.  Recognition of shapes by editing their shock graphs , 2004, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[13]  I. Biederman,et al.  Surface versus edge-based determinants of visual recognition , 1988, Cognitive Psychology.

[14]  Ehud Rivlin,et al.  Local Invariants For Recognition , 1995, IEEE Trans. Pattern Anal. Mach. Intell..

[15]  Vincent Torre,et al.  A versatile segmentation procedure , 2006, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[16]  Farzin Mokhtarian,et al.  Scale-Based Description and Recognition of Planar Curves and Two-Dimensional Shapes , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[17]  Josef Kittler,et al.  Robust and Efficient Shape Indexing through Curvature Scale Space , 1996, BMVC.

[18]  Milan Sonka,et al.  Learning Shape Models from Examples Using Automatic Shape Clustering and Procrustes Analysis , 1999, IPMI.

[19]  Fred L. Bookstein,et al.  Landmark methods for forms without landmarks: morphometrics of group differences in outline shape , 1997, Medical Image Anal..