Search papers, labs, and topics across Lattice.
This paper introduces UnderOneFacade, the largest and most comprehensive benchmark dataset for 3D facade semantic segmentation, featuring 2.7 billion annotated points with centimeter accuracy and harmonized semantic labels. The authors evaluate various architectures, revealing that current state-of-the-art methods struggle with fine-grained architectural recognition and show significant performance degradation across different geographic domains, with the best achieving only 33 IoU on the fine-grained LoFG3 benchmark. By providing a standardized and extensive dataset, UnderOneFacade aims to enhance the development of robust and transferable 3D segmentation models, addressing a critical gap in the field.
Current models falter on fine-grained facade elements, achieving only 33 IoU across geographic domains, highlighting a pressing need for better benchmarks.
Globally consistent semantic digital twins require centimeter-accurate and geographically transferable 3D facade segmentation. However, progress in facade parsing is limited by the lack of large-scale, standardized benchmarks for evaluating cross-domain generalization. Existing datasets are geographically narrow, semantically inconsistent, or insufficiently precise. We introduce UnderOneFacade, the largest cross-country and cross-continent 3D facade benchmark to date, comprising centimeter-accurate point clouds with hierarchical, harmonized, and architecturally grounded semantic labels totaling 2.7 billion annotated points. Through a systematic evaluation of representative point-, graph- and transformer-based architectures, we show that current methods struggle to recognize fine-grained architectural elements and degrade significantly across geographic domains, with the best models achieving only up to 33 IoU on the fine-grained LoFG3 benchmark. By combining geometric precision with standardized semantics at unprecedented scale, UnderOneFacade establishes a rigorous benchmark for developing robust and transferable 3D segmentation models. The dataset, evaluation scripts, and pretrained models will be released upon publication.
