One-shot Compositional Data Generation for Low Resource Handwritten Text Recognition

Mohamed Ali Souibgui, Ali Furkan Biten, Sounak Dey, Alicia Fornes, Yousri Kessentini, Lluis Gomez, Dimosthenis Karatzas, Josep Llados

Producción científica: Capítulo del libroCapítuloInvestigaciónrevisión exhaustiva

9 Citas (Scopus)

Resumen

Low resource Handwritten Text Recognition (HTR) is a hard problem due to the scarce annotated data and the very limited linguistic information (dictionaries and language models). For example, in the case of historical ciphered manuscripts, which are usually written with invented alphabets to hide the message contents. Thus, in this paper we address this problem through a data generation technique based on Bayesian Program Learning (BPL). Contrary to traditional generation approaches, which require a huge amount of annotated images, our method is able to generate human-like handwriting using only one sample of each symbol in the alphabet. After generating symbols, we create synthetic lines to train state-of-the-art HTR architectures in a segmentation free fashion. Quantitative and qualitative analyses were carried out and confirm the effectiveness of the proposed method.

Idioma originalInglés
Título de la publicación alojadaProceedings - 2022 IEEE/CVF Winter Conference on Applications of Computer Vision, WACV 2022
EditorialInstitute of Electrical and Electronics Engineers Inc.
Páginas2563-2571
Número de páginas9
ISBN (versión digital)9781665409155
DOI
EstadoPublicada - 2022

Serie de la publicación

NombreProceedings - 2022 IEEE/CVF Winter Conference on Applications of Computer Vision, WACV 2022

Huella

Profundice en los temas de investigación de 'One-shot Compositional Data Generation for Low Resource Handwritten Text Recognition'. En conjunto forman una huella única.

Citar esto