Summary: Boston University Computer Science Tech. Report No. 2003­009, April 1, 2003.
To appear in Proc. IEEE CVPR, June 2003.
Estimating 3D Hand Pose from a Cluttered Image
Vassilis Athitsos and Stan Sclaroff #
Computer Science Department
Boston University
111 Cummington Street
Boston, MA 02215
email: {athitsos, sclaroff}@cs.bu.edu
Abstract
A method is proposed that can generate a ranked list of
plausible three­dimensional hand configurations that best
match an input image. Hand pose estimation is formulated
as an image database indexing problem, where the closest
matches for an input hand image are retrieved from a large
database of synthetic hand images. In contrast to previ­
ous approaches, the system can function in the presence of
clutter, thanks to two novel clutter­tolerant indexing meth­
ods. First, a computationally efficient approximation of
the image­to­model chamfer distance is obtained by embed­

Source: Athitsos, Vassilis - Department of Computer Science and Engineering, University of Texas at Arlington
Boston University, Department of Computer Science

Collections: Computer Technologies and Information Sciences