Add isotropic Gaussian kernels at own risk: more and more resilient modes in higher dimensions

It has been an open question whether the sum of finitely many isotropic Gaussian kernels in n ≥ 2 dimensions can have more modes than kernels, until in 2003 Carreira-Perpinan and Williams exhibited n+1 isotropic Gaussian kernels in Rn with n+2 modes. We give a detailed analysis of this example, showing that it has exponentially many critical points and that the resilience of the extra mode grows like √n. In addition, we exhibit finite configurations of isotropic Gaussian kernels with superlinearly many modes.

[1]  B. Silverman,et al.  Using Kernel Density Estimates to Investigate Multimodality , 1981 .

[2]  J. Koenderink.  The structure of images , 2004, Biological Cybernetics.

[3]  Stephen J. Roberts,et al.  Parametric and non-parametric unsupervised cluster analysis , 1997, Pattern Recognit..

[4]  Tony Lindeberg,et al.  Scale-Space Theory in Computer Vision , 1993, Lecture Notes in Computer Science.

[5]  Brittany Terese Fasy,et al.  Modes of gaussian mixtures and an inequality for the distance between curves in space , 2012 .

[6]  E M Kohner,et al.  The solution of the problem. , 1978, Transactions of the ophthalmological societies of the United Kingdom.

[7]  S. Sanders,et al.  Solution of Problem , 1929 .

[8]  Feller William,et al.  An Introduction To Probability Theory And Its Applications , 1950 .

[9]  Andrew P. Witkin,et al.  Scale-Space Filtering , 1983, IJCAI.

[10]  Alan L. Yuille,et al.  Scaling Theorems for Zero Crossings , 1987, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[11]  Chao Chen,et al.  Diffusion runs low on persistence fast , 2011, 2011 International Conference on Computer Vision.

[12]  Azriel Rosenfeld,et al.  Computer Vision , 1988, Adv. Comput..

[13]  L. Florack,et al.  The application of catastrophe theory to image analysis , 2001 .

[14]  Christopher K. I. Williams,et al.  An isotropic Gaussian mixture can have more modes than components , 2003 .

[15]  Joachim H. Rieger.  Generic evolutions of edges on families of diffused greyvalue surfaces , 2005, Journal of Mathematical Imaging and Vision.

[16]  LA Vrence.  Segmentation Based on Intensity Extrema , 2004 .

[17]  J. Behboodian.  On the Modes of a Mixture of Two Normal Distributions , 1970 .

[18]  J. Damon.  Local Morse Theory for Solutions to the Heat Equation and Gaussian Blurring , 1995 .

[19]  Andrew P. Witkin,et al.  Uniqueness of the Gaussian Kernel for Scale-Space Filtering , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[20]  Miguel Á. Carreira-Perpiñán,et al.  On the Number of Modes of a Gaussian Mixture , 2003, Scale-Space.