3D point cloud data processing with machine learning for construction and infrastructure applications: A comprehensive review

Abstract

Point clouds are increasingly being used to improve productivity, quality, and safety throughout the life cycle of construction and infrastructure projects. While applicable for visualizing construction projects, point clouds lack meaningful semantic information. Thus, the theoretical benefits of point clouds, such as productivity, quality, and safety improvement, in the construction and infrastructure domains can only be achieved after the processing of point clouds. Manual processing of point cloud datasets is costly, time-consuming, and error-prone. A variety of automatic approaches, such as machine learning methods, are adopted in different steps of automatic processing of point clouds. This article surveys recent research on point cloud datasets, which were automatically processed with machine learning methods in construction and infrastructure industries. An outline for future research is proposed based on identified research gaps. This review paper aims to be a reference for researchers to acknowledge the state-of-the-art applications of automatically-processed point cloud models in construction and infrastructure domains and a guide to assist stakeholders in developing automatic procedures in construction and infrastructure industries.