This is a subproject of the project MobilityLab Stor-Trondheim (MoST) at NTNU. 

In this project, a novel framework will be developed to generate digital twin with detailed geometries, attributes, and semantic information from laser scanning point clouds data. Among others, new deep learning-based algorithms for semantic segmentation and 3D outline extraction will be proposed for traffic related objects (i.e. traffic signs, road marks, urban facilities, etc.) in urban area. This will be integrated together with the existing work for building reconstruction and individual tree detection (ITD) into the framework, so that 3D digital models can be generated with high quality and efficiency.