Principal Investigators 

Prof. Dr.-Ing. André Borrmann, 
Chair of Computational Modeling and Simulation, Department of Civil and Environmental Engineering

Prof. Dr. rer. nat. Daniel Cremers,
Chair of Computer Vision and Artificial Intelligence, Departments of Informatics and Mathematics

Prof. Dr. rer. nat. Thomas H. Kolbe,
Chair of Geoinformatics, Department of Aerospace and Geodesy

Prof. Dr.-Ing. Frank Petzold,
Chair of Architectural Informatics, Department of Architecture

Prof. Dr.-Ing. Ludwig Hoegner,
Photogrammetry and Remote Sensing, Department of Aerospace and Geodesy

Prof. Dr.-Ing. Xiaoxiang Zhu, 
Data Science in Earth Observation, Department of Aerospace and Geodesy

Motivation  and Goals  

Digital modelling of the built word requires information abstractions at various scales and according to a range of competences. This project involves the unique generation of a system of interrelated digital twins of the built environment, spanning multiple resolution scales providing rich semantics and coherent geometry. We are researching a multi-scale, multi-method approach combining terrestrial, airborne, and spaceborne acquisition, with multiple sensor types (visible, thermal, LiDAR, Radar).

Recent Results 

• Automatic pipeline for large-scale building model reconstruction from TomoSAR point clouds.
• Novel top-down-bottom-up method for point clouds processing incorporating construction knowledge.
  Creation of indoor digital twin models for TUM Building 1 at the City Campus.
• Generation of annotated datasets leveraging the learned features of facade elements.
• Registration of TIR point cloud with building model and projection of thermal intensities to façade.
• Automatic pipeline for voxelization of IFC, CityGML and Mesh Models to compare geometry and semantics using RichVoxels as a common representation.

Selected documentation