8 Oct 2025
Three new studies from LMU’s Institute for Digital Cultural Heritage Studies, presented at CIPA 2025, showcase cutting-edge AI, 3D, and VR applications in archaeology.
Researchers at LMU’s Institute for Digital Cultural Heritage Studies (in collaboration with colleagues) have published three papers in The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences stemming from their presentations at CIPA 2025 in Seoul. These publications highlight LMU’s growing role in integrating machine learning, computer vision, and immersive 3D technologies into digital heritage research:
1. Semi-automated LiDAR Vegetation Classification for Mediterranean Archaeology: Designing a Pipeline Leveraging a Multi-Layer Stacked Ensemble Approach presents a hybrid pipeline combining feature engineering, a stacked ensemble classifier, and RandLA-Net to distinguish vegetation from ground in drone-based LiDAR point clouds.
2. StatueDiff: Regionally Enhanced Cross-Scale Diffusion for Single-View 3D Reconstruction of Classical Statues proposes a diffusion-based deep learning framework that inputs a single color photograph and outputs a photorealistic 3D reconstruction of sculptural artefacts, aimed at enhancing museum and digital exhibition workflows.
3. 3D Technologies and Virtual Reality in Archaeology: Preserving the Kition Ship Graffiti from Late Bronze Age Cyprus describes the high-resolution recording (via photogrammetry and terrestrial laser scanning) of ship graffiti on Temple 1’s southern façade, the discovery of 23 previously unknown graffiti, and the creation of an immersive, historically informed virtual reconstruction to support digital preservation and public access.
The three open-access papers appear in ISPRS Archives Volume XLVIII-M-9-2025, linked here: https://isprs-archives.copernicus.org/articles/XLVIII-M-9-2025/