Handwriting Recognition of Historical Documents with Few Labeled Data - Télécom Paris Accéder directement au contenu
Communication Dans Un Congrès Année : 2018

Handwriting Recognition of Historical Documents with Few Labeled Data

Résumé

Historical documents present many challenges for offline handwriting recognition systems, among them, the segmentation and labeling steps. Carefully annotated textlines are needed to train an HTR system. In some scenarios, transcripts are only available at the paragraph level with no text-line information. In this work, we demonstrate how to train an HTR system with few labeled data. Specifically, we train a deep convolutional recurrent neural network (CRNN) system on only 10% of manually labeled text-line data from a dataset and propose an incremental training procedure that covers the rest of the data. Performance is further increased by augmenting the training set with specially crafted multiscale data. We also propose a model-based normalization scheme which considers the variability in the writing scale at the recognition phase. We apply this approach to the publicly available READ dataset 1. Our system achieved the second best result during the ICDAR2017 competition [1].
Fichier principal
Vignette du fichier
published_das2018.pdf (1.47 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04267897 , version 1 (02-11-2023)

Identifiants

Citer

Edgard Chammas, Chafic Mokbel, Laurence Likforman-Sulem. Handwriting Recognition of Historical Documents with Few Labeled Data. 2018 13th IAPR International Workshop on Document Analysis Systems (DAS), Apr 2018, Vienna, France. pp.43-48, ⟨10.1109/DAS.2018.15⟩. ⟨hal-04267897⟩
11 Consultations
4 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More