Creating an information-rich digital twin of indoor environments by interpretation and fusion of image and point-cloud data

A digital twin of built environments can benefit all stakeholders, but its creation process requires high human effort, which makes the cost invested in creating a digital twin exceed the potential benefits it can provide. This dissertation thesis presents an overall approach to creating an information-rich digital twin of buildings by combining object detection methods in point clouds and images. The final result includes large space-bounding elements and small objects with geometric information, object category information, and text information.