An affine invariant deformable shape representation for general curves

Automatic construction of shape models from examples has been the focus of intense research during the last couple of years. These methods have proved to be useful for shape segmentation, tracking and shape understanding. In this paper novel theory to automate shape modelling is described. The theory is intrinsically defined for curves although curves are infinite dimensional objects. The theory is independent of parameterisation and affine transformations. We suggest a method for implementing the ideas and compare it to minimising the description length of the model (MDL). It turns out that the accuracy of the two methods is comparable. Both the MDL and our approach can get stuck at local minima. Our algorithm is less computational expensive and relatively good solutions are obtained after a few iterations. The MDL is, however, better suited at fine-tuning the parameters given good initial estimates to the problem. It is shown that a combination of the two methods outperforms either on its own.

[1]  G. Sparr.  Projective invariants for affine shapes of point configurations , 1991 .

[2]  Dmitry B. Goldgof,et al.  Point correspondence recovery in non-rigid motion , 1992, Proceedings 1992 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[3]  Nicholas Ayache,et al.  Adaptive meshes and nonrigid motion computation , 1994, Proceedings of 12th International Conference on Pattern Recognition.

[4]  Timothy F. Cootes,et al.  Use of active shape models for locating structures in medical images , 1994, Image Vis. Comput..

[5]  Gunnar Sparr,et al.  A Common Framework for Kinetic Depth, Reconstruction and Motion for Deformable Objects , 1994, ECCV.

[6]  Christopher J. Taylor,et al.  Automatic Landmark Generation for Point Distribution Models , 1994, BMVC.

[7]  David C. Hogg,et al.  Learning Flexible Models from Image Sequences , 1994, ECCV.

[8]  G. Sparr.  Structure and motion from kinetic depth , 1995 .

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

[10]  K. Mardia,et al.  Statistical Shape Analysis , 1998 .

[11]  Christopher J. Taylor,et al.  Automatic construction of eigenshape models by direct optimization , 1998, Medical Image Anal..

[12]  Hemant D. Tagare,et al.  Shape-based nonrigid correspondence with application to heart motion analysis , 1999, IEEE Transactions on Medical Imaging.

[13]  Guido Gerig,et al.  Elastic model-based segmentation of 3-D neuroradiological data sets , 1999, IEEE Transactions on Medical Imaging.

[14]  Lawrence H. Staib,et al.  Shape-based 3D surface correspondence using geodesics and local geometry , 2000, Proceedings IEEE Conference on Computer Vision and Pattern Recognition. CVPR 2000 (Cat. No.PR00662).

[15]  Philip N. Klein,et al.  Constructing 2D curve atlases , 2000, Proceedings IEEE Workshop on Mathematical Methods in Biomedical Image Analysis. MMBIA-2000 (Cat. No.PR00737).

[16]  Timothy F. Cootes,et al.  Statistical models of appearance for medical image analysis and computer vision , 2001, SPIE Medical Imaging.

[17]  Timothy F. Cootes,et al.  A minimum description length approach to statistical shape modeling , 2002, IEEE Transactions on Medical Imaging.

[18]  Henrik Aanæs,et al.  Estimation of Deformable Structure and Motion , 2002 .

[19]  Hans Henrik Thodberg,et al.  Minimum Description Length Shape and Appearance Models , 2003, IPMI.

[20]  Kalle Åström,et al.  Extension of Affine Shape , 2004, Journal of Mathematical Imaging and Vision.