From 3D point clouds to HBIM: Application of Artificial Intelligence in Cultural Heritage

Interest in semantic segmentation of 3D point clouds using ML and DL has grown due to their key role in scene insight across a wide range of computer vision, robotics and remote sensing applications. In the domain of Cultural Heritage, 3D point clouds are increasingly used as the backbone for as-built BIM models becoming a conventional approach to design in the AEC industry. However, there's a research gap in this field regarding the interface between point cloud segmentation and the HBIM workflow: there are no consistent studies demonstrating the possibility of automating the construction of parametric historical features from the segmentation process results in terms of geometry and semantic labels. The current research intends to perform a systematic review of the current bibliography with the aim of offering a constructive synthesis that will provide as a springboard for the advancement of innovative strategies in the field of BIM and AI.