Learning Mixtures Of Smooth, Nonuniform Deformation Models For Probabilistic Image Matching

Proceedings of Eighth International Workshop on Artificial Intelligence and Statistics, Key West, FL |

By representing images and image prototypes by linear subspaces spanned by \tangent vectors” (derivatives of an image with respect to translation, rotation, etc.), impressive invariance to known types of uniform distortion can be built into feedforward discriminators. We describe a new probability model that can jointly cluster data and learn mixtures of nonuniform, smooth deformation fields. Our fields are based on low-frequency wavelets, so they use very few parameters to model a wide range of smooth deformations (unlike, e.g., factor analysis, which uses a large number of parameters to model deformations). We give results on handwritten digit recognition and face recognition. Accepted for poster presentation, AISTATS 2001, Key West, FL. 1 Introduction Many computer vision and image processing tasks benefit from invariances to spatial deformations in the image. Examples include handwritten character recognition, face recognition and motion estimation in video sequences.