| | |
Summary: Detecting and Solving Template Ambiguities in Motion Segmentation
Pedro M. Q. Aguiar
Jose M. F. Moura
Instituto de Sistemas e Robotica Electrical and Computer Engineering
Instituto Superior Tecnico Carnegie Mellon University
Lisboa, Portugal Pittsburgh, U.S.A.
E-mail : aguiar@isr.ist.utl.pt E-mail : moura@ece.cmu.edu
Abstract
When the color or gray level of a moving object is
very similar to that of the background, motion-based
segmentation methods fail. This leads to ambiguous
templates for the moving objects. In this paper, we
propose a method that segments unambiguously from
motion the templates of moving objects. Our method,
which we call Incremental Motion Segmentation
integrates over time the small dierences between the
gray level of the moving object and that of the back-
ground. Our experiments with segmenting a robot soc-
cer video clip show the quality of our results.
1 Introduction
|
