Project ISF: Generating LoD3 Building Models for a Digital Twin Program DUIRI - Discovery Undergraduate Interdisciplinary Research Internship Term Summer 2024 Status Accepted Research Area Civil Engineering Description The objective of this project is to establish a framework for generating Level of Detail 3 (LoD3) building models for digital twins. The development of algorithms for 3D building modeling is a well-explored subject in geomatics, serving as a crucial foundation for a wide range of civil engineering industries, including urban planning and transportation management. Currently, the field faces challenges in advancing models from LoD2 to LoD3, which necessitates the inclusion of detailed features such as openings in buildings. To address this, the project employs a variety of deep learning models to extract semantic information from imagery and integrate it into the existing LoD2 models. Through this project, students will gain insights into the general process of 3D building modeling, leveraging artificial intelligence and photogrammetric principles within the field of geomatics. Supervisor Jinha Jung Mentor Jinha Jung Type of work required We are seeking a team of creative minds to generate training data from images collected from drones, develop object detection algorithms (targeting windows and doors from the building) using deep learning models, and collaborate with graduate students from the Geospatial Data Science Lab to integrate the deliverables into the digital twin model. Websites https://gdsl.org Qualifications Students should be in good academic standing. Students are expected to have some experience with programming in Python and machine learning modeling using Scikit-learn, Tensorflow, or similar frameworks. Credit Hours 0 Weekly Hours 10 (estimated)