Building Model
Building model research focuses on creating accurate and detailed 3D representations of structures, primarily for applications in architecture, engineering, construction, and autonomous navigation. Current efforts leverage diverse techniques, including large language models for generating models from textual descriptions, deep learning for semantic segmentation and feature extraction from point clouds and images, and optimization algorithms for refining model geometry and incorporating physical constraints. These advancements improve the efficiency and accuracy of building modeling, enabling better design, analysis, and management of the built environment, and facilitating applications like improved autonomous vehicle localization and energy-efficient building control.
Papers
Enriching thermal point clouds of buildings using semantic 3D building models
Jingwei Zhu, Olaf Wysocki, Christoph Holst, Thomas H. Kolbe
Analyzing the impact of semantic LoD3 building models on image-based vehicle localization
Antonia Bieringer, Olaf Wysocki, Sebastian Tuttas, Ludwig Hoegner, Christoph Holst