WG III/4: 3D Scene Analysis

Our Mission

ISPRS Working group III/4 aims at making progress in the automatic recognition and 3D reconstruction of objects in complex scenes from images, point clouds, and other sensor data. The emphasis is on urban scenes characterized by the occurrence of different object classes such as vegetation, buildings, roads, street furniture and cars, with the goal of generating high-resolution 3D City Models. It is also our goal to evaluate methods for object extraction and the suitability of different sensors for that task.

To this purpose, the WG organises workshops to exchange the latest developments on object recognition and reconstruction. The WG will continue to organize the ISPRS benchmark on urban object detection by adding new data sets and by expanding the current ones.

The WG will organize the review process for a Theme Issue of the ISPRS journal on automated urban object detection.

Finally, the WG will continue to work closely with other WGs, particularly in Commissions III and VII.

Working Group Officers:


Franz Rottensteiner

Franz Rottensteiner
Institute for Photogrammetry and GeoInformation
Leibniz Universität Hannover
Nienburger Str.1
30167 Hannover, Germany
+49 511 762 3893
+49 511 762 2483



Gunho Sohn

Gunho Sohn
GeoICT Lab
Earth and Space Science and Engineering Department
York University
4700 Keele St.
Toronto, M3J 1P3
+1 416 650 8011
+1 416 736 5817



Markus Gerke

Markus Gerke
ITC; EOS Department
University of Twente
Hengelosestraat 99, P.O. Box 6
7500 AA Enschede,
The Netherlands
+31 53 4874 522
+31 53 4874 335



Jan Dirk Wegner

Jan Dirk Wegner
Institute of Geodesy and Photogrammetry
Swiss Federal Institute of Technology Zurich
Wolfgang-Pauli-Strasse 15
8093 Zurich
+41 44 633 68 08
+41 44 633 11 92



Terms of Reference:

  • Models and techniques for extracting objects (buildings, roads, cars, vegetation etc.) from aerial, spaceborne and terrestrial image, laser, and (In)SAR data
  • Generation and update of high-resolution 3D city models and road databases
  • Interpretation of terrestrial, aerial and spaceborne sensor data, possibly together with information from traditional cartographic products, CAD models, and urban GIS.
  • Tools and models for integrating information about multiple object classes and their relations within complex scenes.
  • Automatic and semi-automatic generation of urban models with level-of-detail (LOD) and attributes
  • Analysis of the trade-off between geometry and radiometry / texture for visualization
  • Assessment of efficiency and quality, and of their dependence on the quality of the input data