Development of the Ontology for Preserving 3D Models of Cultural Heritage

Poster presents working in progress of development of ontology of preservation of digital cultural heritage 3D models. 

The graphics show respectively:
- top left: the life cycle of a digital resource
-upper right: proposed docuement scheme for hypothetical digital restorations
-bottom left corner: Attempt to map classes and properties from existing ontologies to the data schema using CIDOC CRM base (grey), CRMdig (green), OntSciDoc3D (blue) and the newly designed onotology (yellow).
-right middle section: descriptions of the ontologies used, from left - CIDOC CRM base, OntScioDoc3D, CRMdig,
-bottom right: hierarchy of classes planned for use with the CIDOC CRM base (white for classes not used in the data model, grey for classes used in the data model), CRMdig (green), OntSciDoc3D (blue) and the new ontology (yellow).

ABSTRACT

In light of recent years, rapid technological developments facilitating access to digital retrodigitisation and reconstruction techniques have brought an overflow of 3D models of digital heritage, challenging the entire community with the need of developing standards to ensure access, traceability and reusability of the data created. Despite the many new approaches to the documentation and publication of 3D models, there is still a lack of shared concepts and definitions when it comes to the details of the scientific approaches to creating digital models [1]. Only proper documentation of the research methods will allow to fully exploit the potential of digital heritage models, which create numerous derivatives used for various presentation and visualisation technologies, such as animations, augmented or virtual reality, 3D printing and light projection, gaming and many others [2].

Initial work was started under the project 'DFG 3D-Viewer. Infrastructure for digital 3D reconstruction'[3], where work began on an analysis of existing 3D model documentation methods, repositories of digital 3D heritage and metadata standards [4]. Collaborations were also established with several scientific projects and initiatives on the documentation and publication of digital heritage in order to exchange experiences and jointly map the documentation schemas used by the various infrastructures [5]. On this basis, a prototype WissKI [6] repository for 3D models was created with use of the OntSciDoc3D ontology [7], testing its relevance in the wider con text of digital heritage. The first tests showed that the used ontology, focusing mainly on the documentation of source-based digital reconstitutions, , pays attention to aspects that work well in a virtual research environment but not in a repository, omitting description of  methods used to create 3D models  and it is not sufficient to address issues related to retrodigitisation models.

It was decided to apply the learned lessons to the new project of Computer-based Visualisation of Architectural Cultural Heritage (CoVHer) with objective of “defining applicative/practical guidelines and operational methodologies aimed at the study, the implementation, the visualization (including access), and the critical evaluation of the 3D models of artefacts that no longer exist or have never been built” [8]. To this end, a work on a new ontology has been initiated to enable low threshold documentation for 3D models of cultural heritage with a focus on preservation and reusability of published data. The  ontology for Preserving 3D models of Cultural Heritage (P3DCH) supposed to be an application ontology based on the CIDOC CRM [9] and its extension CRM Digital (CRMdig) [10] enriched with additional classes and properties focusing on 3D modelling and scientific documentation. The first aspect was to classify the 3D model so that it could be adequately documented. There were three main attributes influencing the way in which it should be documented: the condition of the object influencing the type of conducted research, the methods used to acquire/create the 3D model and the type of object depicted.

Through consultation with the community, it was possible to identify the two main types of research for digital cultural heritage: reality-based retrodigitisation, focusing on techniques mainly related to photogrammetry or laser scanning, and hypothetical source-based reconstruction, strongly related to techniques based on human interpretation of the data [11]. In the case of digital reconstructions, it is necessary to further distinguish between objects that were destroyed and those that were never realised and refer to the reconstruction of the concept depicted on the surviving sources rather than the physical object.

Regarding 3D modelling methods, there is currently a lack of scientific studies indicating a clear division and definitions of the various approaches. Therefore, it is planned to use terms and methods defined in the CoVHer project itself, within which it is expected to prepare a scientific publication containing a “methods and common language/understanding (glossary) for the construction/evaluation of 3D models CVCH (Computer-based Visualisation of Architectural Cultural Heritage)” [8]. The team's discussions to date have already led to the identification of five main modelling methods, which are procedural/algorithm modelling (generating a model through the definition of an ordered set of actions), parametric modelling (generating a model through inputting features and constraints that can be used to automate repetitive changes), reality-based automatic techniques (photogrammetry, laser scanning), handmade direct modelling (generating a model through the use of manual modelling commands that do not establish automatic relationships and constraints) and hybrid methods (a mix of the above methods) [12].

The classification of the digital heritage assets was carried out on the basis of the criteria of the scale of the object (or area). Using the Getty Art & Architecture Thesaurus[13], seven subclasses have been distinguished including: Settlements & Landscapes. Complexes & Districts, Open (public) spaces, Single Built Work, Building Component, Furnishing & Equipment and General Object.

Due to the extensive scope, it was now decided to focus on capturing all aspects related to the source-based digital reconstruction of the architecture by analysing existing ontologies (OntSciDoc3D [14], CHML [15], CRMvr [16]) and evaluating the scientific value of model development according to the guidelines presented in the Scientific Reference Model (SRM)[17] and Critical Digital Model (CRM) [18] methodologies. The second element to be implemented is the preparation of the ontology for capturing the retrodigitisation process using existing ontology developments in this field [19–22].

BIBLIORAPHY

1.            Kuroczyński, P.: Virtual Research Environment for digital 3D reconstructions – Standards, thresholds and prospects. Studies in Digital Heritage. 1, 456–476 (2017). https://doi.org/10.14434/sdh.v1i2.23330

2.            Muenster, S.: Digital 3D Technologies for Humanities Research and Education: An Overview. Applied Sciences. 12, 2426 (2022). https://doi.org/10.3390/app12052426

3.            DFG-Viewer: DFG 3D-Viewer, https://dfg-viewer.de/en/dfg-3d-viewer

4.            Bajena, I., Kuroczyński, P., Münster, S.: Metadata scheme for 3D re-constructions. How to capture the source–based 3D reconstruction of Cultural Heritage? (2021)

5.            Metadata Workshop in Mainz, https://miro.com/app/board/uXjVPDUEGMY=

6.            Wiss-KI.eu, http://wiss-ki.eu/about

7.            OntSciDoc3D, https://www.ontscidoc3d.hs-mainz.de/ontology/

8.            Computer-based Visualization of Architectural Cultural Heritage, https://covher.eu/

9.            CIDOC CRM, https://www.cidoc-crm.org/

10.         CRMdig, https://cidoc-crm.org/crmdig/

11.         Münster, S., Friedrichs, K., Hegel, W.: 3D Reconstruction Techniques as a Cultural Shift in Art History? 23 (2018). https://doi.org/10.11588/dah.2018.3.32473

12.         Fallavollita, F.: Introduction to Computer-based Visualization of Architectural Cultural Heritage. CoVHer Multiplyer Event. , Mainz (2023)

13.         Art & Architecture Thesaurus (Getty Research Institute), https://www.getty.edu/research/tools/vocabularies/aat/

14.         Kuroczynski, P., Große, P.: OntSciDoc3D – Ontology for Scientific Documentation of source-based 3D reconstruction of architecture. Presented at the The 25th Conference on Cultural Heritage and New Technologies , Vienna, Austria (2020)

15.         Kuroczyński, P.: CULTURAL HERITAGE MARKUP LANGUAGE –DESIGNING A DOMAIN ONTOLOGY FOR DIGITAL RECONSTRUCTIONS. VIRTUAL ARCHAEOLOGY (Methods and benefits). (2015)

16.         Giovannini, E.C.: CRMvr, https://github.com/elisabettacaterina/CRMvr, (2018)

17.         Piotr Kuroczyński, Igor Bajena, Irene Cazzaro: Scientific Reference Model. A methodological approach in hypothetical 3D reconstruction of architecture. Presented at the The 27th Conference on Cultural Heritage and New Technologies , Vienna, Austria (2022)

18.         Apollonio, F.I., Fallavollita, F., Foschi, R.: The Critical Digital Model for the Study of Unbuilt Architecture. In: Research and Education in Urban History in the Age of Digital Libraries. pp. 3–24. Springer, Cham (2021)

19.         Amico, N., Felicetti, A.: Ontological Entities for Planning and Describing Cultural Heritage 3D Models Creation. (2021)

20.         Bruseker, G., Guillem, A., Carboni, N.: Semantically Documenting Virtual Reconstruction: Building a Path to Knowledge Provenance. ISPRS Annals of Photogrammetry, Remote Sensing and Spatial Information Sciences. II-5/W3, 33–40 (2015). https://doi.org/10.5194/isprsannals-II-5-W3-33-2015

21.         Homburg, T., Cramer, A., Raddatz, L., Mara, H.: Metadata schema and ontology for capturing and processing of 3D cultural heritage objects. Heritage Science. 9, 91 (2021). https://doi.org/10.1186/s40494-021-00561-w

22.         D’Andrea, A., Fernie, K.: CARARE 2.0: A metadata schema for 3D cultural objects. In: 2013 Digital Heritage International Congress (DigitalHeritage). pp. 137–143 (2013)