1
Return

FP-THD: Full page transcription of historical documents

delete2026-07-04
delete0
delete
OA
AI
H
Hala Neji *
J
Javier Nogueras-Iso
J
Javier Lacasta
M
Miguel Ángel Latre
F
Francisco Javier García-Marco
DOI：10.1016/j.patcog.2026.114351delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
• Proposal of pipeline for full page image transcription specialized in medieval Latin. • Integration and customization of models for layout analysis and OCR. • Evaluation with existing datasets and new dataset for Latin transcription. • Effective performance for printed and handwritten transcription.
Keywords:
Masked-autoencoder
Deep learning
Vision transformer
Historical document

Journal

Pattern Recognition cover
Pattern Recognition
IF:
7.6
Papers:
1.3W
Citations:
4.5W

Organization

U
universidad de zaragoza
Scholars:
498
Papers: 181
Citations: 0
Cited Papers

Cited Papers

Citing Papers

Citing Papers