Semantic models of the road spaces in the inner city of Ingolstadt in Germany at the level of detail 3:
- the comprehensive model of the road spaces is provided as OGC CityGML 2.0 and 3.0 datasets
- the road network is provided as an ASAM OpenDRIVE demo dataset from the company 3D Mapping Solutions
Download the semantic models by cloning the repo:
git clone --depth 1 git@github.com:savenow/lod3-road-space-models.git- models/road-network/opendrive
- Content: Road network with roadside objects
- Source: Demo dataset by the company 3D Mapping Solutions
- Conceptual data model: ASAM OpenDRIVE 1.4
- Encoding: XML
- License:
CC BY-NC-SA 4.0
- models/road-network/citygml
- Content: Road network with roadside objects
- Source: Converted from OpenDRIVE demo dataset to CityGML 2.0 and 3.0 using the tool r:trån
- Conceptual data model: OGC CityGML 2.0 and 3.0
- Encoding: GML
- License:
CC BY-NC-SA 4.0
- models/building/lod3
- Content: Over 50 LOD3 building models
- Source: Manually modeled based on the MLS point clouds by 3D Mapping Solutions
- Conceptual data model: OGC CityGML 2.0 and 3.0
- Encoding: GML
- License:
CC BY-SA 4.0
- models/building/lod2
- Content: 4 tiles of LOD2 building models
- Source: Open data portal of the Bavarian State Mapping Agency
- Conceptual data model: OGC CityGML 2.0 and 3.0
- Encoding: GML
- License:
CC BY 4.0
The LOD3 building models were manually modeled using the tool SketchUp with the CityEditor extension. For this, mobile laser scanning (MLS) point clouds by the company 3D Mapping Solutions GmbH with a relative accuracy of 1-3 cm served as the basis for modeling. The SketchUp project files are also provided along with a creation guideline that documents the entire modeling process.
Further information:
- Modeling aspects: What is modeled?
- Overview: Web-Map-Client with the LOD3 building and road network models (kudos to Christof Beil)
To expand and improve the dataset, feedback and contributions are always appreciated. You can also contact me directly via mail benedikt.schwab@tum.de.
Very special acknowledgments are due to Sophie Haas Goschenhofer and Olaf Wysocki, who worked diligently on the development of the methodology and the realization of the modeling.