Tiling of Ortho-Rectified Façade Images

Przemyslaw Musialski, Meinrad Recheis, Stefan Maierhofer, Peter Wonka, Werner Purgathofer
Tiling of Ortho-Rectified Façade Images
In Proceedings of 26th Spring Conference on Computer Graphics (SCCG 2010). May 2010.
[

draft]

Content:

Information
Abstract
Additional Files and Images
Weblinks
BibTeX

Information

Publication Type: Conference Paper
Workgroup(s)/Project(s):
- VRVis Competence Center
Date: May 2010
Booktitle: Proceedings of 26th Spring Conference on Computer Graphics (SCCG 2010)
Date (from): 13. May 2010
Date (to): 15. May 2010
Editor: Helwig Hauser , Reinhard Klein
ISBN: 978-80-223-2644-5
Lecturer: Przemyslaw Musialski
Location: Budmerice, Slovak Republic
Publisher: Comenius University, Bratislava
Keywords: image processing, facade segmentation, urban reconstruction

Abstract

Typical building facades consist of regular structures such as windows arranged in a predominantly grid-like manner. We propose a method that handles precisely such facades and assumes that there must be horizontal and vertical repetitions of similar patterns. Using a Monte Carlo sampling approach, this method is able to segment repetitive patterns on orthogonal images along the axes even if the pattern is partially occluded. Additionally, it is very fast and can be used as a preprocessing step for finer segmentation stages.

Additional Files and Images

Additional images and videos

image

Additional files

draft

Weblinks

No further information available.

BibTeX

@inproceedings{musialski-2010-tof,
  title =      "Tiling of Ortho-Rectified Fa\c{c}ade Images",
  author =     "Przemyslaw Musialski and Meinrad Recheis and Stefan
               Maierhofer and Peter Wonka and Werner Purgathofer",
  year =       "2010",
  abstract =   "Typical building facades consist of regular structures such
               as windows arranged in a predominantly grid-like manner. We
               propose a method that handles precisely such facades and
               assumes that there must be horizontal and vertical
               repetitions of similar patterns. Using a Monte Carlo
               sampling approach, this method is able to segment repetitive
               patterns on orthogonal images along the axes even if the
               pattern is partially occluded. Additionally, it is very fast
               and can be used as a preprocessing step for finer
               segmentation stages.",
  month =      may,
  booktitle =  "Proceedings of 26th Spring Conference on Computer Graphics
               (SCCG 2010)",
  editor =     "Helwig Hauser , Reinhard Klein",
  isbn =       "978-80-223-2644-5",
  location =   "Budmerice, Slovak Republic",
  publisher =  "Comenius University, Bratislava",
  keywords =   "image processing, facade segmentation, urban reconstruction",
  URL =        "https://www.cg.tuwien.ac.at/research/publications/2010/musialski-2010-tof/",
}