Summary: Properties of Patch Based Approaches for the
Recognition of Visual Object Classes
Alexandra Teynor1
, Esa Rahtu2
, Lokesh Setia1
, and Hans Burkhardt1
1
University of Freiburg, Department of Computer Science
Georges K¨ohler Allee 052, 79110 Freiburg, Germany
{teynor, setia, Hans.Burkhardt}@informatik.uni-freiburg.de
2
University of Oulu, Department of Electrical and Information Engineering,
PO Box 4500, 90014 Oulu, Finland
erahtu@ee.oulu.fi
Abstract. Patch based approaches have recently shown promising re-
sults for the recognition of visual object classes. This paper investigates
the role of different properties of patches. In particular, we explore how
size, location and nature of interest points influence recognition perfor-
mance. Also, different feature types are evaluated. For our experiments
we use three common databases at different levels of difficulty to make

Source: Albert-Ludwigs-Universität Freiburg, Institut für Informatik,, Lehrstuhls für Mustererkennung und Bildverarbeitung

Collections: Computer Technologies and Information Sciences