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 ,. 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....
KeywordsIEEE Transaction Unsupervised Cluster Shape Context Procrustes Analysis Curvature Scale Space
Unable to display preview. Download preview PDF.
- 4.Edelman, S. (ed.): Representation and Recognition in Vision. MIT Press, Cambridge (1999)Google Scholar
- 6.Coggins, J.: A statistical approach to multiscale, medial vision. Technical report, Department of Computer Science, University of North Carolina at Chapel Hill (1992)Google Scholar
- 7.Blum, H.: A transformation for extracting new descriptors of shape. In: Wathen-Dunn, W. (ed.) Models for the Perception of Speech and Visual Form, pp. 362–380. MIT Press, Cambridge, USA (1967)Google Scholar
- 8.Sebastian, T., Klien, P., Kimia, B.: Recognition of shapes by editing their shock graphs. IEEE Transaction on PAMI 26, 550–571 (2004)Google Scholar
- 9.Mokhtarian, F., Mackworth, A.: Scale-based description and recognition of planar curves and two-dimensional shapes. IEEE Transaction on PAMI 8, 34–43 (1986)Google Scholar
- 10.Belongie, S., Malik, J.: Shape matching and object recognition using shape contexts. IEEE Transaction on PAMI 24, 509–522 (2002)Google Scholar
- 11.Arbter, K., Snyder, W., Burhardt, H., Hirzinger, G.: Application of affine-invariant fourier descriptors to recognition of 3-d objects. IEEE Transaction on PAMI 12, 640–647 (1990)Google Scholar
- 12.Rivlin, E., Weiss, I.: Local invariants for recognition. IEEE Transaction on PAMI 17, 226–238 (1995)Google Scholar
- 14.Duta, N., Sonka, M., Jain, A.: Learning shape models from examples using automatic shape clustering and procrustes analysis. LNCS. Springer, Heidelberg (1999)Google Scholar
- 15.Srivastava, A., Joshi, S., Mio, W., Liu, X.: Statistical shape analysis: Clustering, learning, and testing. IEEE Transaction on PAMI 27, 590–602 (2005)Google Scholar
- 19.Mokhtarian, F., Abbasi, S., Kittler, J.: Efficient and robust shape retrieval by shape content through curvature scale space. In: Proc. First Int’l Conf. Image Database and Multi-Search, pp. 35–42 (1996)Google Scholar
- 20.Vanzella, W., Torre, V.: A versatile segmentation procedure. IEEE Transactions on SMC, Part B 36, 366–378 (2006)Google Scholar