What question did this study set out to answer?

This research aims to develop a point cloud-free workflow for accurately reconstructing LOD2 building models from RGB orthophotos.

CityZen: LOD2 building reconstruction with a point cloud-free model-driven approach

View Full Paper

Ask AI

Bookmark

Authors

Randomized trial demonstrates scalable LOD2 city modeling in urban environments, suggesting a cost-effective solution.

This research aims to develop a point cloud-free workflow for accurately reconstructing LOD2 building models from RGB orthophotos.
Integrated monocular height estimation using DSMNet, HTC-DC-Net, and TSE-Net.
Roof type classification and model-driven reconstruction in a unified pipeline.
Use of building footprints as geometric constraints for 3D structure generation.
Achieved accurate 3D building models from RGB orthophotos without relying on costly 3D data.
Demonstrated effective integration of height estimation and structure generation, improving scalability.
Enabled practical city-scale modeling, reducing dependency on traditional data acquisition methods.

Büyükdemircioğlu et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark