2025 EC3 & CIB W78 - Krause, Nils Sören - A Foundation Model Approach: Generating Point Clouds ... скачать в хорошем качестве

2025 EC3 & CIB W78 - Krause, Nils Sören - A Foundation Model Approach: Generating Point Clouds ...

Title: A Foundation Model Approach: Generating Point Clouds for Scan-to-BIM Training and BIM Model Generation Authors: Krause, Nils Sören; Ploennig, Joern Affiliation: Universität Rostock, Germany Keywords: Synthetic Data Generation, Foundation Models, Scan-to-BIM, Outlier Elimination Session: Data Sensing, Acquisition and Analysis Paper Link: www.dx.doi.org/ 10.35490/EC3.2025.243 Abstract: Modern scan-to-BIM approaches often use machine learning (ML) models to reconstruct BIM models from point clouds. Their performance relies heavily on a large amount of training data. Generating this data requires a significant amount of manual effort for outlier elimination and labelling. The approach uses a foundation model (FM) for the generation of point cloud training data. The presented generator enables the training of AI algorithms for the scan-to-BIM method without additional hardware. It also allows future researchers to generate data for a scan-to-BIM workflow.