Authors:
Roman Martel
1
;
Chaoqun Dong
1
;
Kan Chen
1
;
Henry Johan
2
and
Marius Erdt
3
Affiliations:
1
Fraunhofer Singapore and Singapore
;
2
Nanyang Technological University, Fraunhofer IDM@NTU and Singapore
;
3
Fraunhofer Singapore, Singapore, Nanyang Technological University, Fraunhofer IDM@NTU and Singapore
Keyword(s):
Computer Vision, 3D Building Models, City Area Maps, Text Recognition, Deep Learning.
Related
Ontology
Subjects/Areas/Topics:
Applications and Services
;
Computer Vision, Visualization and Computer Graphics
;
Image and Video Analysis
;
Imaging for Cultural Heritage (Modeling/Simulation, Virtual Restoration)
;
Segmentation and Grouping
Abstract:
In this paper, we propose a pipeline that converts buildings described in city area maps to 3D models in the CityGML LOD1 standard. The input documents are scanned city area maps provided by a city authority. The city area maps were recorded and stored over a long time period. This imposes several challenges to the pipeline such as different font styles of typewriters, handwritings of different persons, varying layout, low contrast, damages and scanning artifacts. The novel and distinguishing aspect of our approach is its ability to deal with these challenges. In the pipeline we, firstly, identify and analyse text boxes within the city area maps to extract information like height and location of its described buildings. Secondly, we extract the building shapes based on these locations from an online city map API. Lastly, using the extracted building shapes and heights, we generate 3D models of the buildings.