A Fuzzy 3D Registration Method
Abstract
An important problem in computer vision is to recover how features
extrated from images are connected to either existing models. In this paper, we
focus on solving the registration problem, i.e obtaining rigid displacement
parameters between several 3D data, whether partial or exhaustive. We present
a general method performing robust 3D location and fitting based on fuzzy
clustering method handling noisy motions. We show results on syntetic and
real 3D data.
Reference
@INPROCEEDINGS{jpteufit95,
author = {Tarel, Jean-Philippe and Boujemaa, Nozha},
title = {A Fuzzy 3{D} Registration Method},
booktitle = {Proceedings of Third European Congress Intelligent Techniques and Soft Computing (EUFIT'95)},
address = {Aachen, Germany},
date = {28-31 August},
pages = {1359-1365},
year = {1995},
note = {http://perso.lcpc.fr/tarel.jean-philippe/publis/eufit95.html}
}
Pdf file (429 Kb)
(c) EUFIT