Machine-Learning-Enhanced Procedural Modeling for 4D Historical Cities Reconstruction

The generation of 3D models depicting cities in the past holds great potential for documentation and educational purposes. However, it is often hindered by incomplete historical data specialized expertise required. To address these challenges, we propose a framework city reconstruction. By integrating procedural modeling techniques machine learning within Geographic Information System (GIS) framework, our pipeline allows effective management spatial detailed models. We developed an open-source Python module that fills gaps 2D GIS datasets directly generates up to LOD 2.1 from files. use CityJSON format ensures interoperability accommodates specific needs A practical case study using footprints Old City Jerusalem between 1840 1940 demonstrates creation, completion, representation dataset, highlighting versatility effectiveness approach. This research contributes accessibility accuracy models, providing tools informative incorporating maintaining dynamic nature ensure possibility supporting ongoing updates refinement based on newly acquired data. Our methodology offers streamlined solution reconstruction, eliminating need additional software increasing usability practicality process.