3D Urban Generation and Surface Reconstruction

The field of 3D urban generation and surface reconstruction is moving towards more realistic and detailed models, with a focus on geometry-aware and appearance-controllable methods. Researchers are exploring new approaches to address the limitations of existing methods, such as the need for large-scale 3D city assets and the reliance on semantic or height maps. Noteworthy papers include Sat2RealCity, which proposes a framework for 3D urban generation from satellite imagery, and SparseSurf, which introduces a method for sparse-view 3D Gaussian Splatting for surface reconstruction. SF-Recon is also notable for its simplification-free lightweight building reconstruction via 3D Gaussian Splatting.

Sources

Sat2RealCity: Geometry-Aware and Appearance-Controllable 3D Urban Generation from Satellite Imagery

SF-Recon: Simplification-Free Lightweight Building Reconstruction via 3D Gaussian Splatting

SparseSurf: Sparse-View 3D Gaussian Splatting for Surface Reconstruction

Built with on top of