Assessing indoor spaces in existing buildings has recently evolved beyond conventional measurement techniques. Contemporary practices now involve point cloud utilizing laser scanners, enabling precise spatial data capture. This thesis delves into transforming segmented point cloud data into parametric 3D models. This transformation is accomplished via a structured framework. The primary objective of this research is to harness the potential of segmented point cloud data for parametric modeling of indoor space. This modeling process plays a crucial role when the functionality of a building undergoes changes or when there is a need to enhance facility management workflows and achieve additional objectives. The input of the proposed framework includes 3D semantic segmented and space-wise labeled point clouds and the adjacency matrices among corresponding spaces. The outcome of the study consists of parametric models, which provide a detailed representation of the indoor environment, such as space dimensions, height, boundary thickness, and lengths. The proposed method is tested and validated on different datasets. The results show the effectiveness of the proposed method in creating a parametric model. The average accuracy of the constructed walls in the constructed model and the reference model is 0.13 m in length and less than 0.1 m in width, and height.     «

Assessing indoor spaces in existing buildings has recently evolved beyond conventional measurement techniques. Contemporary practices now involve point cloud utilizing laser scanners, enabling precise spatial data capture. This thesis delves into transforming segmented point cloud data into parametric 3D models. This transformation is accomplished via a structured framework. The primary objective of this research is to harness the potential of segmented point cloud data for parametric modeling o...     »