This paper addresses the need for a more streamlined and automated pipeline for generating 3D models of Cultural Heritage buildings, particularly churches, for acoustic simulations. Traditional methods often rely heavily on manual reconstruction from point cloud sections, which is time-consuming and prone to inaccuracies. To overcome these limitations, this research introduces a novel modeling pipeline that enhances efficiency while maintaining geometric accuracy. The methodology begins with the acquisition of point clouds through Terrestrial Laser Scanning (TLS), followed by an optimized point cloud processing and mesh generation workflow. CloudCompare is utilized for cleaning and surface reconstruction using advanced algorithms such as Delaunay Triangulation and Poisson Surface Reconstruction. The generated meshes are further refined in Blender, where operators like decimation, smoothing, and remeshing improve quality while reducing complexity. A verification step ensures the mesh is watertight and free from defects, making it suitable for acoustic simulations. The pipeline was tested on the Church of San Flaviano in Giulianova, Italy, demonstrating that this approach produces accurate models in a significantly shorter time compared to traditional methods. The primary results indicate that the proposed methodology successfully delivers a high-quality, simulation-ready model with improved efficiency. Future research will focus on comparing the models obtained from the new methodology and the traditional one, and integrating a material layer system to enhance acoustic accuracy, ensuring broader applicability in architectural research and Cultural Heritage conservation.
Proposing a New 3D Modeling Pipeline for Churches Geometrical Acoustic Simulations: The Case Study of San Flaviano
Ricciutelli, Alessandro
;Sangiorgio, Valentino
2025-01-01
Abstract
This paper addresses the need for a more streamlined and automated pipeline for generating 3D models of Cultural Heritage buildings, particularly churches, for acoustic simulations. Traditional methods often rely heavily on manual reconstruction from point cloud sections, which is time-consuming and prone to inaccuracies. To overcome these limitations, this research introduces a novel modeling pipeline that enhances efficiency while maintaining geometric accuracy. The methodology begins with the acquisition of point clouds through Terrestrial Laser Scanning (TLS), followed by an optimized point cloud processing and mesh generation workflow. CloudCompare is utilized for cleaning and surface reconstruction using advanced algorithms such as Delaunay Triangulation and Poisson Surface Reconstruction. The generated meshes are further refined in Blender, where operators like decimation, smoothing, and remeshing improve quality while reducing complexity. A verification step ensures the mesh is watertight and free from defects, making it suitable for acoustic simulations. The pipeline was tested on the Church of San Flaviano in Giulianova, Italy, demonstrating that this approach produces accurate models in a significantly shorter time compared to traditional methods. The primary results indicate that the proposed methodology successfully delivers a high-quality, simulation-ready model with improved efficiency. Future research will focus on comparing the models obtained from the new methodology and the traditional one, and integrating a material layer system to enhance acoustic accuracy, ensuring broader applicability in architectural research and Cultural Heritage conservation.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.


