English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Reconstructing semantic 3D building models at the level of detail (LoD) 3 isa long-standing challenge. Unlike mesh-based models, they require watertightgeometry and object-wise semantics at the fa\c{c}ade level. The principalchallenge of such demanding semantic 3D reconstruction is reliablefa\c{c}ade-level semantic segmentation of 3D input data. We present a novelmethod, called Scan2LoD3, that accurately reconstructs semantic LoD3 buildingmodels by improving fa\c{c}ade-level semantic 3D segmentation. To this end, weleverage laser physics and 3D building model priors to probabilisticallyidentify model conflicts. These probabilistic physical conflicts proposelocations of model openings: Their final semantics and shapes are inferred in aBayesian network fusing multimodal probabilistic maps of conflicts, 3D pointclouds, and 2D images. To fulfill demanding LoD3 requirements, we use theestimated shapes to cut openings in 3D building priors and fit semantic 3Dobjects from a library of fa\c{c}ade objects. Extensive experiments on the TUMcity campus datasets demonstrate the superior performance of the proposedScan2LoD3 over the state-of-the-art methods in fa\c{c}ade-level detection,semantic segmentation, and LoD3 building model reconstruction. We believe ourmethod can foster the development of probability-driven semantic 3Dreconstruction at LoD3 since not only the high-definition reconstruction butalso reconstruction confidence becomes pivotal for various applications such asautonomous driving and urban simulations.

Scan2LoD3: Reconstructing semantic 3D building models at LoD3 using ray casting and Bayesian networks