Beijing Institute of Mathematical Sciences and Applications

Chenliang Wang , Zhaoyuan Yu

Luojian Tan

Sunday, November 3, 2024 7:00 PM - 8:00 PM

A7-201

Tencent 809 329 731 ()

3D urban building modeling is vital foundational data for building digital twins and creating smart cities. In response to existing challenges such as high time costs, complex production processes, and low consistency with real-world textures in large-scale 3D urban building modeling (3UBM) methods, this article proposes a "rebuild 3D urban building models" approach that integrates geospatial data and street view. The approach achieves enhanced generation of large-scale city-level LOD2 3UBM. Based on open-source OSM data, SVI data, and other resources, the approach was tested in modeling experiments conducted in Hong Kong, Shanghai, and other cities. A dataset covering 30 cities in China was constructed to demonstrate the approach's characteristics of large coverage area, high time efficiency, high model quality, and low economic cost. The accuracy of texture mapping reached 85%, and the modeling cost was only one-twentieth of that of traditional methods, with modeling efficiency 87 times higher. This achievement has significant economic value in bridging the gap in the production of large-scale and low-cost 3UBM data, promoting the construction of smart cities, CIM, and real-world 3D modeling.