Communication Dans Un Congrès Année : 2024

Normalized vs Diplomatic Annotation: A Case Study of Automatic Information Extraction from Handwritten Uruguayan Birth Certificates

Solène Tarride
Seginus Mowlavi
Antoine Tadros
Gabriele Facciolo
Christopher Kermorvant
Javier Preciozzi

Résumé

This study evaluates the recently proposed Document Attention Network (DAN) for extracting key-value information from Uruguayan birth certificates, handwritten in Spanish. We investigate two annotation strategies for automatically transcribing handwritten documents, fine-tuning DAN with minimal training data and annotation effort. Experiments were conducted on two datasets containing the same images (201 scans of birth certificates written by more than 15 different writers) but with different annotation methods. Our findings indicate that normalized annotation is more effective for fields that can be standardized, such as dates and places of birth, whereas diplomatic annotation performs much better for fields containing names and surnames, which can not be standardized.

Fichier principal
Vignette du fichier
ADAPDA_2024_paper___OCR_Normalized_vs__Diplomatic_Annotation.pdf (1.23 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Licence

Dates et versions

hal-04769409 , version 1 (08-04-2025)

Licence

Identifiants

Citer

Natalia Bottaioli, Solène Tarride, Jérémy Anger, Seginus Mowlavi, Marina Gardella, et al.. Normalized vs Diplomatic Annotation: A Case Study of Automatic Information Extraction from Handwritten Uruguayan Birth Certificates. Document Analysis and Recognition – ICDAR 2024 Workshops, Aug 2024, Athens, Greece. pp.40-54, ⟨10.1007/978-3-031-70645-5_4⟩. ⟨hal-04769409⟩
188 Consultations
191 Téléchargements

Altmetric

Partager

  • More