A semi-automatic 2D/3D annotation framework for the geometric analysis of heritage artefacts - Archive ouverte HAL Access content directly
Conference Papers Year : 2018

A semi-automatic 2D/3D annotation framework for the geometric analysis of heritage artefacts

Abstract

Documentation and monitoring of heritage objects involve many actors on multidisciplinary aspects. The progress made over the years in the field of digital technologies has enabled many tools for analysis, management and dissemination of information gathered around an object. These tools must allow users to semantically describe the object while allowing them to grasp its morphological complexity and the heterogeneity of the available analysis supports. This article introduces an approach for the semantic annotation of heritage objects by using the bijective relationship that can be established between a 3D representation of an object and the set of oriented images towards it, while maintaining a continuum of information between all phases of observation and description, from acquisition to visualization of semantically enriched representations. The main idea is to offer a versatile environment to help extraction of relevant information from images using geometric descriptors and semi-automatic point cloud processing methods.
Fichier principal
Vignette du fichier
A semi-automatic 2D3D annotation framework for the geometric analysis of heritage artefacts.pdf (3.39 Mo) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-02326770 , version 1 (22-10-2019)

Identifiers

Cite

Adeline Manuel, Anas Alaoui M'Darhri, Violette Abergel, Fabien Rozar, Livio De Luca. A semi-automatic 2D/3D annotation framework for the geometric analysis of heritage artefacts. 3rd Digital Heritage International Congress (Digital Heritage) held jointly with 2018 24th International Conference on Virtual Systems & Multimedia (VSMM 2018), Oct 2018, San Francisco, United States. pp.1-7, ⟨10.1109/DigitalHeritage.2018.8810114⟩. ⟨hal-02326770⟩
113 View
172 Download

Altmetric

Share

Gmail Facebook X LinkedIn More