In this paper we proposed a methodology that describes the major steps of a scan-to-BIM process. The methodology includes six steps: (1) classification of considered elements, (2) definition of required level of detail (GI), (3) scan data acquisition, (4) point cloud registration and segmentation, (5) as-built BIM creation and (6) analysis. The examples of the application of the proposed methodology are demonstrated by creation of as-built BIM models for existing industrial sites and historic buildings. As the results of these case studies have shown, the proposed methodology can be used for as-built BIMs without any prior information.

Laser scanning is an essential method for monitoring of the operation of buildings or structures. It involves creating as-is BIM from point clouds obtained from laser scanning. In this article we present our workflow for the generation of information model from 3D point clouds of concrete tetrapod blocks on navigable structure C-1. Point cloud processing method for making informational model for long term monitoring is described. As a result of the research BIM model with each tetrapod was created for deformational monitoring in the comparison with next year model. Finally, we identify and discuss technology gaps that need to be addressed in future research.

At the present time, many publicly available point cloud datasets exist, which are mainly focused on autonomous driving. The objective of this study is to develop a new large-scale mobile 3D LiDAR point cloud dataset for outdoor scene semantic segmentation tasks, which has a classification scheme suitable for geospatial applications. Our dataset (Saint Petersburg 3D) contains both real-world (34 million points) and synthetic (34 million points) subsets that were acquired using real and virtual sensors with the same characteristics. An original classification scheme is proposed that contains a set of 10 universal object categories into which any scene represented by dense outdoor mobile LiDAR point clouds can be divided. The evaluation procedure for semantic segmentation of point clouds for geospatial applications is described. An experiment with the Kernel Point Fully Convolution Neural Network model trained on the proposed dataset was carried out. We obtained an overall 92.56% mIoU, which demonstrates the high efficiency of using deep learning models for point cloud semantic segmentation for geospatial applications in accordance with the proposed classification scheme.