From geometric rules to ribbed vaults procedural modeling. Parametric tools to study star vaults geometry in St-Eustache

The definition of adaptive parametric models to automate modelling architectural elements is one of the main goal for heritage documentation, dissemination and management in HBIM process. Parametric objects library of historical architectural elements is a very important step in more than one process, starting from 3D Point Clouds segmentation with Deep Learning (DL) techniques to 3D modelling from Point Cloud. Following the state of the art we have defined an adaptive parametric model for ribbed vaults modelling that allow us to generate a 3D model based on geometric rules and\or from point cloud data.. Starting from geometric rules we defined the main parameters to use to generate different “ideal models” and the specific parameters from point cloud data to use to generate a “reality based model”. We compared the different 3D models based on geometric rules from literature, 3D reality based model with mesh model from point cloud data to evaluate the process, to improve it and to verify the design hypothesis. We are going to test this workflow on a case study: Saint-Eustache church in Paris. There are some different kind of star vaults in Saint-Eustache, in this paper we show the parametric model definition process from French literature for the five keystone vault samples.