Hasso-Plattner-Institut
  
Hasso-Plattner-Institut
Prof. Dr. Jürgen Döllner
  
 
Rico Richter

Contact

Tel.: +49-(0)331 5509-3910
Fax: +49-(0)331 5509-172
Room: H-2.26
E-Mail: rico.richter(at)hpi.de

Publikationen

Object class segmentation of massive 3D point clouds of urban areas using point cloud topology

Richter, Rico and Behrens, Markus and Döllner, Jürgen
International Journal of Remote Sensing, vol. 34(23):8408-8424 2013

DOI: DOI: 10.1080/01431161.2013.838710

Abstract:

A large number of remote-sensing techniques and image-based photogrammetric approaches allow an efficient generation of massive 3D point clouds of our physical environment. The efficient processing, analysis, exploration, and visualization of massive 3D point clouds constitute challenging tasks for applications, systems, and workflows in disciplines such as urban planning, environmental monitoring, disaster management, and homeland security. We present an approach to segment massive 3D point clouds according to object classes of virtual urban environments including terrain, building, vegetation, water, and infrastructure. The classification relies on analysing the point cloud topology; it does not require per-point attributes or representative training data. The approach is based on an iterative multi-pass processing scheme, where each pass focuses on different topological features and considers already detected object classes from previous passes. To cope with the massive amount of data, out-of-core spatial data structures and graphics processing unit (GPU)-accelerated algorithms are utilized. Classification results are discussed based on a massive 3D point cloud with almost 5 billion points of a city. The results indicate that object-class-enriched 3D point clouds can substantially improve analysis algorithms and applications as well as enhance visualization techniques.

BibTeX file

@article{RKD12,
author = { Richter, Rico and Behrens, Markus and Döllner, Jürgen },
title = { Object class segmentation of massive 3D point clouds of urban areas using point cloud topology },
journal = { International Journal of Remote Sensing },
year = { 2013 },
volume = { 34 },
number = { 23 },
pages = { 8408-8424 },
month = { 0 },
abstract = { A large number of remote-sensing techniques and image-based photogrammetric approaches allow an efficient generation of massive 3D point clouds of our physical environment. The efficient processing, analysis, exploration, and visualization of massive 3D point clouds constitute challenging tasks for applications, systems, and workflows in disciplines such as urban planning, environmental monitoring, disaster management, and homeland security. We present an approach to segment massive 3D point clouds according to object classes of virtual urban environments including terrain, building, vegetation, water, and infrastructure. The classification relies on analysing the point cloud topology; it does not require per-point attributes or representative training data. The approach is based on an iterative multi-pass processing scheme, where each pass focuses on different topological features and considers already detected object classes from previous passes. To cope with the massive amount of data, out-of-core spatial data structures and graphics processing unit (GPU)-accelerated algorithms are utilized. Classification results are discussed based on a massive 3D point cloud with almost 5 billion points of a city. The results indicate that object-class-enriched 3D point clouds can substantially improve analysis algorithms and applications as well as enhance visualization techniques. },
priority = { 0 }
}

Copyright Notice

last change: Mon, 12 Jan 2015 12:02:38 +0100

Vorträge

  • "Detection and Classification of Changes in Urban Areas Based on Semantic Analysis of Airborne Point Clouds", European LiDAR Mapping Forum 2014, Amsterdam, The Netherlands (12/2014) 
  • "Out-of-Core Visualization of Classified 3D Point Clouds", 3D GeoInfo 2014, Dubai, VAE (11/2014) 
  • "Effiziente Bestandsaktualisierung von 3D-Stadtmodellen durch Analyse multi temporaler 3D-Punktwolken", Workshop 3D-Stadtmodelle 2014, Bonn, Deutschland (11/2014) 
  • "Semantische Klassifizierung von 3D-Punktwolken für Stadtgebiete", Terrestrisches Laserscanning 2013, Fulda, Deutschland (12/2013) 
  • "Potentiale von massiven 3D-Punktwolkendatenströmen", Geoinformatik 2012, Braunschweig, Deutschland (03/2012) 
  • "Ein Ansatz für die Differenzanalyse zwischen 3D-Punktwolken und 3D-Referenzgeometrie", DGPF - Jahrestagung 2011, Mainz, Deutschland (04/2011)
  • "Software-Technologie für massive 3D-Punktwolken", Laserscanning –eine Methode zur Bereitstellung von Geodaten für vielfältigste Anwendungen am GFZ Potsdam, Deutschland (04/2011)
  • "Bestandsaktualisierung von 3D-Stadtmodellen durch Analyse von 3D-Punktwolken", auf der 3-Ländertagung 2010, Wien , Deutschland (07/2010)
  • "Out-of-Core Real-Time Visualization of Massive 3D Point Clouds", auf der 7th International Conference on Virtual Reality, Computer Graphics, Visualisation and Interaction in Africa, Franschoek, South Africa (06/2010)

Lehrtätigkeiten

Wintersemester 2014/2015

  •  Seminar Geovisualisierung
  •  Bachelorprojekt "3D-Punktwolken: Big Spatial Data"

Sommersemester 2014

  •  Seminar Information Visualization
  •  Bachelorprojekt "Analyse & Visualisierung von 3D-Punktwolken"

Wintersemester 2013/2014

  •  Seminar Geovisualisierung
  •  Bachelorprojekt "Analyse & Visualisierung von 3D-Punktwolken"

Sommersemester 2013

  •  Seminar Geovisualisierungsverfahren

Wintersemester 2012/2013

  •  Seminar Konzepte und Techniken der 3D-Visualisierung

Sommersemester 2012

  •  Seminar Geovisualisierungsverfahren

Wintersemester 2011/2012

  •  Seminar Echtzeit-Rendering-Techniken

Sommersemester 2011

  •  Seminar Geovisualisierung

Wintersemester 2010/2011

  • Tutor Vorlesung Informationsvisualisierung
  • Seminar Advanced Visualization Techniques

Sommersemester 2010

  • Bachelorprojekt "Werkzeug zur Analyse umfangreicher Laser-Scans"

Wintersemester 2009/2010

  • Tutor Vorlesung Geovisualisierung
  • Bachelorprojekt "Werkzeug zur Analyse umfangreicher Laser-Scans"

Sommersemester 2009

  • Übung zur Vorlesung 3D Computergrafik