7th German GeoForum and 25th anniversary event of DDGI; [7. deutsches geoforum und jubiläumsveranstaltung 25 jahre DDGI]

Abstract

This article discusses point clouds and artificial intelligence and explores how intelligent methods can be used to process and interpret three-dimensional spatial data. Point clouds are increasingly produced by laser scanning, photogrammetry, mobile mapping, and other surveying technologies. Their high level of detail provides valuable information about objects and surfaces but also creates challenges for storage, processing, classification, and interpretation. Artificial intelligence can assist by learning patterns within point-cloud datasets and supporting automated identification of different objects or surface types. The article considers the potential of machine-learning approaches for making point-cloud processing more efficient and reducing repetitive manual tasks. AI can contribute to classification, segmentation, feature extraction, and other analytical steps within a geospatial workflow. The quality of the results depends on the characteristics of the point cloud, the training information, and the selected methods. The article highlights the importance of combining computational techniques with professional knowledge so that automated results can be evaluated appropriately. As three-dimensional data becomes more common in surveying, mapping, construction, and infrastructure management, efficient interpretation methods will become increasingly important. The discussion shows how AI can add value to point-cloud workflows and support the transition toward more automated processing of detailed spatial information.