Main objective of the proposed research is to improve accuracy and efficiency of surveying operations, ultimately paving the way for novel 3-D project delivery workflows. This requires the development of novel classification tools for the supervised segmentation of 3-dimensional point clouds by combining digital models, a comprehensive database of laser-based reconstructions, and computer vision libraries such as OpenCV, Scikit-Image, and TorchVision. We will take a Supervised Learning approach in the delivery of the project, to directly include expert knowledge in the classification algorithms. This will be achieved by classifying available point clouds and using this information to devise infrastructure-specific segmentation algorithms. The ultimate objective is the preparation of a report and a set of guidelines to maximize outreach and direct applicability of the research.