PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Learning nonlinear image manifolds by global alignment of local linear models
Jakob Verbeek
IEEE Transactions on Pattern Analysis & Machine Intelligence Volume 28, Number 8, pp. 1236-1250, 2006.

Abstract

Appearance based methods, based on statistical models of the pixels values in an image (region) rather than geometrical object models, are increasingly popular in computer vision. In many applications the number of degrees of freedom (DOF) in the image generating process is much lower than the number of pixels in the image. If there is a smooth function that maps the DOF to the pixel values, then the images are confined to a low dimensional manifold embedded in the image space. We propose a method based on probabilistic mixtures of factor analyzers to (i) model the density of images sampled from such manifolds and (ii) recover global parameterizations of the manifold. A globally non-linear probabilistic two-way mapping between coordinates on the manifold and images is obtained by combining several, locally valid, linear mappings. We propose a parameter estimation scheme that improves upon an existing scheme, and experimentally compare the presented approach to self-organizing maps, generative topographic mapping, and mixtures of factor analyzers. In addition, we show that the approach also applies to find mappings between different embeddings of the same manifold.

EPrint Type:Article
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Machine Vision
Learning/Statistics & Optimisation
ID Code:4220
Deposited By:Jakob Verbeek
Deposited On:04 December 2008