Example files for specific CityGML classes

Sample CityGML files




A basic procedural modelling engine for generating random (synthetic) buildings and other features in CityGML in multiple LODs. The engine is composed of two modules. The first one is procedural: it randomly generates buildings and their elements according to a set of rules and constraints. The buildings are realised through parameters which are stored in an XML form. The second part of the engine reads this data and constructs CityGML data in multiple LODs.

The software and the generated data are free to use, but read this notice


Several Dutch cities


All the urban features (terrain, water, roads, buildings, vegetations, etc.) are present. Datasets were reconstructed with 3dfier by combining the 2D topographic dataset with LiDAR altimetry datasets

The software and the generated data are free to use.


First v2.0 CityGML file


This is the first publicly available CityGML v2.0 data set. It is a 3D model demonstrating nearly all CityGML feature classes from the CityGML 2.0 specification. Especially instances of the classes from the new thematic modules Bridge and Tunnel are included. SolitaryVegetationObjects (the trees) are using ImplicitGeometries with local coordinates to allow for reuse of the geometry definitions. The features also have old and new thematic attributes. Please note that the dataset has been created manually and represents an artificial scenery. Coordinates are not georeferenced and the scene is downscaled to around 1:220. The dataset is decomposed into 6 separate files.

Karl-Heinz Häfele, Research Center Karlsruhe, Institute for Applied Computer Science


Three LOD3 Buildings (CityGML v1.0)


This CityGML data set displays three LoD3 Building objects including the following feature types: GroundSurface, WallSurface, RoofSurface, BuildingInstallation, Window, Door. In addition, CityFurniture and SolitaryVegetation Object features have been modelled on the basis of CityGML’s implicit geometry representation providing three levels of detail. The data set also includes a Road object that consists of various TrafficArea as well as AuxiliaryTrafficArea features. Considering the number of modelled objects and their detailed representation, the memory footprint of this CityGML file is relatively small. This is due to the usage of implicit geomety representations and the efficient data export from SketchUp™, which does not require face triangulation.

GEORES: all data were generated using GEORES tools for SketchUp