SVI2LoD3: Agent-Driven Reconstruction of LoD3 Façade Openings in Semantic 3D City Models from Volunteered Street View Imagery using Large Language and Visual Models
An end-to-end, agent-driven pipeline for the LoD3 reconstruction of façade openings in 3D city models, producing directly usable CityGML-conform outputs and a novel evaluation metric for façade reconstruction, termed Facade Feature Distance.