Image-to-Image Translation Approach for Page Layout Analysis and Artificial Generation of Historical Manuscripts - Centre Jean Mabillon
Chapitre D'ouvrage Année : 2024

Image-to-Image Translation Approach for Page Layout Analysis and Artificial Generation of Historical Manuscripts

Résumé

Document layout analysis is essential in Optical Character Recognition (OCR) and Handwritten Text Recognition (HTR), especially for historical and low-resource scripts. This study explores a novel data augmentation technique using Generative Adversarial Networks (GANs) to generate realistic document layouts from semantic masks, enhancing layout analysis without increasing human annotation effort. Our lightweight pipeline, tested on historical manuscripts (Latin, Arabic, Armenian, Hebrew), newspapers, and complex document layouts, shows that GAN-generated layouts are convincing and difficult to distinguish from real ones, even for paleographers. This method significantly boosts data augmentation, yielding a 3% point improvement in layout analysis metrics (precision, recall, mAP), and a 12 point increase in precision and recall for damaged documents. Additionally, masks with character information enhance image quality, boosting text recognition performance.
Fichier principal
Vignette du fichier
main.pdf (16.69 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
licence

Dates et versions

hal-04707440 , version 1 (24-09-2024)

Licence

Identifiants

Citer

Chahan Vidal-Gorène, Jean-Baptiste Camps. Image-to-Image Translation Approach for Page Layout Analysis and Artificial Generation of Historical Manuscripts. Document Analysis and Recognition – ICDAR 2024 Workshops, 14936, Springer Nature Switzerland, pp.140-158, 2024, Lecture Notes in Computer Science, ⟨10.1007/978-3-031-70642-4_9⟩. ⟨hal-04707440⟩
37 Consultations
19 Téléchargements

Altmetric

Partager

More