Inconsistency of Template Estimation by Minimizing of the Variance/Pre-Variance in the Quotient Space

We tackle the problem of template estimation when data have been randomly deformed under a group action in the presence of noise. In order to estimate the template, one often minimizes the variance when the influence of the transformations have been removed (computation of the Fr{\'e}chet mean in the quotient space). The consistency bias is defined as the distance (possibly zero) between the orbit of the template and the orbit of one element which minimizes the variance. In the first part, we restrict ourselves to isometric group action, in this case the Hilbertian distance is invariant under the group action. We establish an asymptotic behavior of the consistency bias which is linear with respect to the noise level. As a result the inconsistency is unavoidable as soon as the noise is enough. In practice, template estimation with a finite sample is often done with an algorithm called "max-max". In the second part, also in the case of isometric group finite, we show the convergence of this algorithm to an empirical Karcher mean. Our numerical experiments show that the bias observed in practice can not be attributed to the small sample size or to a convergence problem but is indeed due to the previously studied inconsistency. In a third part, we also present some insights of the case of a non invariant distance with respect to the group action. We will see that the inconsistency still holds as soon as the noise level is large enough. Moreover we prove the inconsistency even when a regularization term is added.

[1]  B. Charlier Necessary and sufficient condition for the existence of a Fréchet mean on the circle , 2013 .

[2]  Jean Meunier,et al.  Average Brain Models: A Convergence Study , 2000, Comput. Vis. Image Underst..

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

[4]  Wen Cheng,et al.  Bayesian Registration of Functions and Curves , 2013, 1311.2105.

[5]  Leo Grady,et al.  Spectral Log-Demons: Diffeomorphic Image Registration with Very Large Deformations , 2013, International Journal of Computer Vision.

[6]  K. Mardia,et al.  A penalized likelihood approach to image warping , 2001 .

[7]  Huaiyu Zhu On Information and Sufficiency , 1997 .

[8]  Y. Amit,et al.  Towards a coherent statistical framework for dense deformable template estimation , 2007 .

[9]  Guido Gerig,et al.  Unbiased diffeomorphic atlas construction for computational anatomy , 2004, NeuroImage.

[10]  Théodore Papadopoulo,et al.  Jitter-Adaptive Dictionary Learning - Application to Multi-Trial Neuroelectric Signals , 2013, ICLR.

[11]  Wei Wu,et al.  Signal Estimation Under Random Time-Warpings and Nonlinear Signal Alignment , 2011, NIPS.

[12]  Timothy F. Cootes,et al.  Groupwise Diffeomorphic Non-rigid Registration for Automatic Model Building , 2004, ECCV.

[13]  H. Ziezold On Expected Figures and a Strong Law of Large Numbers for Random Elements in Quasi-Metric Spaces , 1977 .

[14]  Guido Gerig,et al.  Morphometry of anatomical shape complexes with dense deformations and sparse parameters , 2014, NeuroImage.

[15]  J. Tukey,et al.  An algorithm for the machine calculation of complex Fourier series , 1965 .

[16]  Xavier Pennec,et al.  Template Shape Estimation: Correcting an Asymptotic Bias , 2016, SIAM J. Imaging Sci..

[17]  Stephan Huckemann,et al.  Inference on 3D Procrustes Means: Tree Bole Growth, Rank Deficient Diffusion Tensors and Perturbation Models , 2010, 1002.0738.

[18]  F James Rohlf,et al.  Bias and error in estimates of mean shape in geometric morphometrics. , 2003, Journal of human evolution.

[19]  Alain Trouvé,et al.  Template Estimation in Computational Anatomy: Fréchet Means Top and Quotient Spaces Are Not Consistent , 2016, SIAM J. Imaging Sci..

[20]  Xavier Pennec,et al.  Inconsistency of Template Estimation with the Fréchet Mean in Quotient Space , 2017, IPMI.

[21]  R. Bhattacharya,et al.  Statistics on Riemannian manifolds: asymptotic distribution and curvature , 2008 .

[22]  Alain Trouvé,et al.  Computing Large Deformation Metric Mappings via Geodesic Flows of Diffeomorphisms , 2005, International Journal of Computer Vision.

[23]  Anuj Srivastava,et al.  Parameterization-Invariant Shape Statistics and Probabilistic Classification of Anatomical Surfaces , 2011, IPMI.

[24]  D'arcy W. Thompson,et al.  On Growth and Form , 1917, Nature.

[25]  Victor M. Panaretos,et al.  Amplitude and phase variation of point processes , 2016, 1603.08691.