Summary: Boston University Computer Science Tech. Report No. 2005021, June 13, 2005.
Revised February 17, 2006.
To appear in Proc. of the European Conference on Computer Vision (ECCV), May 2006.
Detecting Instances of Shape Classes That Exhibit
Vassilis Athitsos 1 , Jingbin Wang 2 , Stan Sclaroff 2 , and Margrit Betke 2
1 Integrated Data Systems Department
Siemens Corporate Research
755 College Road East
Princeton, NJ 08540, USA
2 Computer Science Department
111 Cummington Street
Boston, MA 02215, USA
Abstract. This paper proposes a method for detecting shapes of variable struc
ture in images with clutter. The term ``variable structure'' means that some shape
parts can be repeated an arbitrary number of times, some parts can be optional,
and some parts can have several alternative appearances. The particular variation
of the shape structure that occurs in a given image is not known a priori. Ex
isting computer vision methods, including deformable model methods, were not