Summary: A Novel Approach to Depth Ordering in Monocular Image Sequences
L. Bergen
Institute for Pattern Recognition and Image Processing
Computer Science Department, University of Freiburg
Am Flughafen 17, D­79110 Freiburg, Germany
bergen@informatik.uni­freiburg.de
F. Meyer
Centre de Morphologie Math’ ematique
Ecole des Mines de Paris
35, rue Saint­Honor’ e
F­77305 Fontainebleau, France
meyer@cmm.ensmp.fr
Abstract
Object oriented representation of image sequences requires
accurate motion segmentation and depth ordering tech­
niques. Unfortunately, the lack of precise motion estimates
at the object boundaries makes these two tasks very difficult.
In this paper we present a detailed analysis of the behaviour
of dense motion estimation techniques at object boundaries
which reveals the systematic nature of the motion estimation

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

Collections: Computer Technologies and Information Sciences