docExtractor: An off-the-shelf historical document element extraction - ANR - Agence nationale de la recherche Accéder directement au contenu
Communication Dans Un Congrès Année : 2020

docExtractor: An off-the-shelf historical document element extraction

Tom Monnier
  • Fonction : Auteur
  • PersonId : 1090223
Mathieu Aubry

Résumé

We present docExtractor, a generic approach for extracting visual elements such as text lines or illustrations from historical documents without requiring any real data annotation. We demonstrate it provides high-quality performances as an off-the-shelf system across a wide variety of datasets and leads to results on par with state-of-the-art when fine-tuned. We argue that the performance obtained without fine-tuning on a specific dataset is critical for applications, in particular in digital humanities, and that the line-level page segmentation we address is the most relevant for a general purpose element extraction engine. We rely on a fast generator of rich synthetic documents and design a fully convolutional network, which we show to generalize better than a detection-based approach. Furthermore, we introduce a new public dataset dubbed IlluHisDoc dedicated to the fine evaluation of illustration segmentation in historical documents.
Fichier principal
Vignette du fichier
paper_final.pdf (3.35 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03131070 , version 1 (04-02-2021)

Identifiants

Citer

Tom Monnier, Mathieu Aubry. docExtractor: An off-the-shelf historical document element extraction. 2020 17th International Conference on Frontiers in Handwriting Recognition (ICFHR), Sep 2020, Dortmund, France. pp.91-96, ⟨10.1109/ICFHR2020.2020.00027⟩. ⟨hal-03131070⟩
52 Consultations
84 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More