Metadata Schema for 3D Data Publication and Archiving - Université de Rennes Accéder directement au contenu
Poster De Conférence Année : 2023

Metadata Schema for 3D Data Publication and Archiving

Schéma de métadonnées pour la publication et l'archivage de données 3D

Résumé

The French National 3D Data Repository, the C.I.N.E.S (National Computer Center for Higher Education) which offers long-term archiving of electronic data, and aLTAG3D - their companion software to document the deposits, use a unique and dedicated metadata scheme based on the work from 2014 to 2017 of the national consortium “3D for Humanities” from Huma-Num infrastructure. This first version is currently evolving to take into account the evolution of deposited data and research domains. Scientific Domains. The first scheme was focused on archaeology - the main domain in digital humanities that make use of 3D models. One of our goals is to offer such previously cited services to the entire Humanities and Social Sciences. For this purpose, we divide metadata into generic and domain specific ones, such as 3D restitutions or scans of “physical object” in archaeology. Automatic Extraction. The metadata are also categorized in two groups: those which can be extracted automatically for the data themselves (such as number of points in a 3D file, creation date from EXIF), and those which have to be filled manually. The automatic process simplifies the user work, and the remaining information volume is still quite large, and we are working on either making it smaller or easier to fill. For this purpose, we are pushing forward the experience gained in developing and experimenting the first scheme. Referencing data. While data redundancy is a guarantee of sustainability, this has an economic and environmental impact. This is even more important for our 3D deposit since, for an open science perspective, we recommend integrating all the documentation, independently of the format and the origin. To tackle this problem, we are integrating into the metadata scheme the possibility to outsource data curation, by using URI to refer to any data present of a sufficiently perennial and reliable infrastructure. We favor institutional hosting over private enterprise. Corresponding requested metadata has also to be harvested. This process is becoming easier with the development of FAIR and institutional infrastructures. Standards. The FAIR principles also impose the interoperability of our scheme with standard formats. For this purpose, we ensure the compatibility with Dublin Core, CIDOC CRM and with DOI. We also integrate the use of standard thesaurus such as PeriodO, GeoNames, PACTOLS. 3D data definition. The structure of our metadata scheme has evolved due to our experience in understanding 3D data and their usage. Initially, our 3D object was only defined by a colored or textured mesh. However, it seemed important to include any kind of 3D, from volumetric data to CAO (used in the history of technologies). We need also to consider how to take into account the object's structure, appearance properties (reflection, etc.), animations, annotations, and everything that can be attached to a 3D object. Furthermore, the 3D data are intrinsically linked with the software and the hardware that have been used to process it and/or to interact with (e.g., in VR usage). We thus extend the metadata scheme in order to include them.
Fichier principal
Vignette du fichier
Metadata Schema for 3D Data Publication and Archiving.pdf (786.13 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04115874 , version 1 (02-06-2023)

Licence

Paternité

Identifiants

  • HAL Id : hal-04115874 , version 1

Citer

Valentin Grimaud, Sylvie Eusèbe, Sarah Tournon, Mehdi Chayani, Matthieu Quantin, et al.. Metadata Schema for 3D Data Publication and Archiving. DARIAH Annual Event 2023: Cultural Heritage Data as Humanities Research Data?, Jun 2023, Budapest, Hungary. ⟨hal-04115874⟩
200 Consultations
20 Téléchargements

Partager

Gmail Facebook X LinkedIn More