Work

Exploring Point Transformers on 3D Semantic Segmentation of Javanese Architectures

Year: 2026

Type: article

Abstract: . The complex geometry of Javanese architecture poses significant challenges for 3D semantic segmentation in cultural heritage documentation. This study evaluates state-of-the-art Point Transformers, i.e., PTv1, PTv2, PTv3, and LitePT, on the Sewu temple dataset, focusing on robustness and efficiency. While PTv1 and PTv2 achieve the highest Intersection-over-Union (mIoU 0.71), they incur high computational costs. Conversely, LitePT provides an optimal balance, delivering competitive results (0.69 mIoU) while being drastically faster. Furthermore, experiments with limited data reveal the significant benefits of transfer learning from European heritage datasets. We conclude that efficient Point Transformer architectures are promising for the automated understanding of complex non-European monuments.

Source: The international archives of the photogrammetry, remote sensing and spatial information sciences/International archives of the photogrammetry, remote sensing and spatial information sciences

DOI: 10.5194/isprs-archives-xlviii-2-w12-2026-57-2026

Authors: Thodoris Betsas, Arnadi Murtiyoso, Pierre Grussenmeyer, Andreas Georgopoulos

Institutions: Centre National de la Recherche Scientifique, National Technical University of Athens, Institut National des Sciences Appliquées de Strasbourg, Laboratoire des Sciences de l'Ingénieur, de l'Informatique et de l'Imagerie

Language: English

Topic: 3D Surveying and Cultural Heritage

Subfield: Geology

Field: Earth and Planetary Sciences

Domain: Physical Sciences

SDG: Industry, innovation and infrastructure

Open Access status: diamond

OpenAlex ID: W7128742045