Automatic retro-structuration of auction sales catalogs layout and content

This paper showcases a pipeline for automatically retro-structuring auction sales catalogs, based on document layout analysis and information extraction technologies. Structured layout and textual data are then transformed into TEI XML for publication. It also advocates for a generalized use of layout segmentation in digitization pipelines.

Mots clés

Machine Learning Document layout analysis Information Extraction TEI Digital Editions

Domaines

Intelligence artificielle [cs.AI] Art et histoire de l'art

Fichier principal

DH2024_DataCatalogue_hscheithauer_sbeniere_lromary.pdf (3.59 Mo)

DH2024_DataCatalogue_hscheithauer_sbeniere_lromary_slides.pdf (3.87 Mo)

Origine	Fichiers produits par l'(les) auteur(s)

Hugo Scheithauer : Connectez-vous pour contacter le contributeur

https://hal.science/hal-04547239

Soumis le : lundi 15 avril 2024-16:16:57

Dernière modification le : mercredi 16 octobre 2024-14:42:09

Dates et versions

hal-04547239 , version 1 (15-04-2024)

Licence

Paternité

Identifiants

HAL Id : hal-04547239 , version 1

Citer

Hugo Scheithauer, Sarah Bénière, Laurent Romary. Automatic retro-structuration of auction sales catalogs layout and content. DH2024 - Reinvention and Responsibility, Alliance of Digital Humanities Organizations, Aug 2024, Washinghton DC, United States. ⟨hal-04547239⟩

Exporter

BibTeX XML-TEI Dublin Core DC Terms EndNote DataCite

Collections

EPHE INRIA INRIA2 PSL

331 Consultations

370 Téléchargements