AUTOMATA - Reference enriched 3D data (D5.3)
Authors/Creators
Description
This deliverable presents the main results of Work Package 5 and the implementation of a software-supported workflow for generating enriched 3D models of archaeological artefacts. The principal result of D5.3 consists of the software components and reference datasets made available with this document, while the report provides the accompanying methodological and technical documentation. Building upon the ontology and metadata framework (D5.1) and the dedicated 3D database (D5.2), the work delivered in D5.3 includes the creation of reference enriched datasets and the methods used to register archaeometric measurements with high-resolution 3D models.
A key software component of this workflow is the adaptation of the Reference Information System in 3D (RIS3D) to the requirements of the AUTOMATA project. RIS3D provides an interactive platform for visualising, managing and annotating enriched 3D models, enabling hyperspectral images, pXRF and Raman measurements to be spatially linked to the 3D surface. The materials and code created for the implementation of this deliverable are available in openly accessible repositories:
(i)the RIS3D application, together with the initial collection of enriched 3D models and associated multimodal datasets, can be downloaded and accessed through the AUTOMATA RIS3D platform at https://automata.3dris.science/. Access and installation instructions are provided in Section 4.2
(ii) the code developed for the automatic selection and visualisation of analytical target locations is available at https://github.com/minstudio/automata_crosshair.
The source code of RIS3D and the AUTOMATA-specific software adaptations will be available with Deliverable 6.2 Visualisation and Interaction on M28.
The document also includes the automated workflows developed to integrate multimodal archaeometric data into enriched 3D models. These include an automated photogrammetric reconstruction pipeline, hyperspectral image registration, and the automatic detection of Regions of Interest (ROIs) to support the selection of target locations for pXRF and Raman analyses. Together with the calibration monitoring, quality assessment, and adaptive acquisition framework presented in D4.1, these developments establish the methodological foundations for the AUTOMATA closed-loop digitisation workflow.
Section 1 introduces the document and outlines the objectives of this deliverable. Section 2 describes the photogrammetric acquisition setup and parameters. Section 3 presents the photogrammetric and hyperspectral registration workflow, which automatically aligns hyperspectral images with the 3D model. This is a key step in the creation of enriched 3D models. Section 4 introduces the Reference Information System in 3D (RIS3D), explains how it operates, and provides guidance on how to use it to access and explore enriched 3D models. Section 5 describes the automatic detection of Regions of Interest (ROIs) for XRF and Raman analyses. Finally, Section 6 concludes the document and outlines future developments related to the registration and visualisation of enriched 3D models.
Files
D5.3 Reference enriched 3D data.pdf
Additional details
Funding
References
- Bleyer, M., Rhemann, C., & Rother, C. (2011). PatchMatch stereo: Stereo matching with slanted support windows. In BMVC, 11, 1–11.
- Campello, R. J. G. B., Moulavi, D., Zimek, A., & Sander, J. (2015). Hierarchical Density Estimates for Data Clustering, Visualization, and Outlier Detection. ACM Transactions on Knowledge Discovery from Data, 10(1), 1–51
- Cernea, D. (2020). OpenMVS: Multi-View Stereo Reconstruction Library (Version 2.4.0) [GitHub Repository]. https://cdcseacave.github.io
- Dutailly, B., Portais, J.-C., & Granier, X. (2023). RIS3D: A Referenced Information System in 3D. Journal of Computing and Cultural Heritage, 15, 73:1–73:20. https://doi.org/10.1145/3517043
- Fei, B., Xu, J., Zhang, R., Zhou, Q., Yang, W., & He, Y. (2024). 3D Gaussian splatting as new era: A survey. IEEE Transactions on Visualization and Computer Graphics.
- Lowe, D. G. (1999). Object recognition from local scale-invariant features. In Proceedings of the Seventh IEEE International Conference on Computer Vision, 2, 1150–1157.
- Maté-González, M. Á., Yali, R., Rodríguez-Hernández, J., González-González, E., & Aguirre de Mata, J. (2025). Comparison of NeRF- and SfM-based methods for point cloud reconstruction for small-sized archaeological artifacts. Remote Sensing, 17(14), 2535.
- Matl, M. (2019). Pyrender (Version 0.1.45) [GitHub Repository]. https://github.com/mmatl/pyrender
- McInnes, L., Healy, J., & Astels, S. (2017). hdbscan: Hierarchical density based clustering. Journal of Open Source Software, 2(11), 205. https://doi.org/10.21105/joss.00205
- Schönberger, J. L., & Frahm, J. M. (2016). Structure-from-Motion Revisited. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 4104–4113.
- Umeyama, S. (1991). Least-squares estimation of transformation parameters between two point patterns. IEEE Transactions on Pattern Analysis and Machine Intelligence, 13(4), 376–380.
- Wang, F., Zhu, Q., Chang, D., Gao, Q., Han, J., Zhang, T., ... & Pollefeys, M. (2026). Learning-based multi-view stereo: A survey. IEEE Transactions on Pattern Analysis and Machine Intelligence, 48, 6241–6260.
- Yao, M., Huo, Y., Ran, Y., Tian, Q., Wang, R., & Wang, H. (2026). Neural radiance field-based visual rendering: A comprehensive review. IEEE Transactions on Visualization and Computer Graphics.
- Zheng, P., Gao, D., Fan, D. P., Liu, L., Laaksonen, J., Ouyang, W., & Sebe, N. (2024). Bilateral Reference for High-Resolution Dichotomous Image Segmentation. CAAI Artificial Intelligence Research, 3, 9150038.