Idiap Higher-order Statistics in Visual Object Recognition

In this paper we develop a higher order statistical theory of matching models against images The basic idea is not only to take into account how much of an object can be seen in the image but also what parts of it are jointly present We show that this additional information can improve the speci city i e reduce the probability of false positive matches of a recognition algorithm We demonstrate formally that most commonly used quality of match measures employed by recognition algorithms are based on an independence assumption Using the Minimum Description Length MDL principle and a simple scene description language as a guide we show that this independence assumption is not satis ed for common scenes and propose several important higher order statistical properties of matches that approximate some aspects of these statistical dependencies We have implemented a recognition system that takes advantage of this additional statistical information and demon strate its e cacy in comparisons with a standard recognition system based on bounded error matching We also observe that the existing use of grouping and segmentation meth ods has signi cant e ects on the performance of recognition systems that are similar to those resulting from the use of higher order statistical information Our analysis provides a statistical framework in which to understand the ef fects of grouping and segmentation on recognition and suggests ways to take better advantage of such information