Points2Model: A Neural-Guided 3D Building Wireframe Reconstruction from Airborne LiDAR Point Clouds
Published in International Journal of Digital Earth, 2025
This paper proposes a neural-guided pipeline for generating 3D building wireframe models directly from airborne LiDAR point clouds. The method tackles the key challenges of façade and roof occlusions, point density variations, and sensor noise inherent to aerial acquisition. By integrating point upsampling and completion into the wireframe reconstruction process, the approach produces accurate, complete building models suitable for urban-scale digital earth applications.
Recommended citation: Akwensi, P. H., Bharadwaj, A., & Wang, R. (2025). "Points2Model: A Neural-Guided 3D Building Wireframe Reconstruction from Airborne LiDAR Point Clouds." International Journal of Digital Earth. 18(1), 2458682.
Download Paper