An important feature of many problem domains in machine learning is their geometry. For example, adjacency relationships, symmetries, and Cartesian coordinates are essential to an...
We propose a methodology for anonymization of records by aggregating sequential records. This methodology is applied to real valued, binary, nominal and ordinal data. We propose tw...
This paper presents a new approach for multi-view object class detection. Appearance and geometry are treated as separate learning tasks with different training data. Our approach...
We propose using the proximity distribution of vectorquantized local feature descriptors for object and category recognition. To this end, we introduce a novel "proximity dis...
Object modeling under consideration of geometric constraints is an important task. In this paper we describe a novel approach to achieving this goal. It has the nice property that...