A Document Attention Network (DAN) model dedicated to Information Extraction from 19th century French Land Registry tables

Fuente: Zenodo
Salvato in:
Dettagli Bibliografici
Autori principali: Tual, Solenn, Abadie, Nathalie, Chazalon, Joseph, Duménieu, Bertrand, Perret, Julien
Natura: Recurso digital
Lingua:francese
Pubblicazione: Zenodo 2025
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866901605246304256
author Tual, Solenn
Abadie, Nathalie
Chazalon, Joseph
Duménieu, Bertrand
Perret, Julien
author_facet Tual, Solenn
Abadie, Nathalie
Chazalon, Joseph
Duménieu, Bertrand
Perret, Julien
contents <p>This repository contains the weights of a Document Attention Network (DAN) (Coquenet et al.) model fine-tuned to perform extraction information from digitised tables of the 19th century French Land Registry. </p> <p>The dataset used to train and evaluate the model is available on Zenodo : <a href="https://zenodo.org/records/15411507" rel="nofollow">10.5281/zenodo.15411507</a>. Annotation model is described in the dataset repository. </p> <p>Be aware that this model has been trained with images from the Departement of Valde-Marne. It may not suceed to extract informations from land registry documents from other departements because of other layouts, local languages, etc.</p> <p><a href="https://gitlab.teklia.com/atr/dan">atr-dan</a> Python library has been used to train the model.</p> <p>DAN architecture as been presented in the following paper : "Denis Coquenet, Clément Chatelain, Thierry Paquet, <a href="https://arxiv.org/abs/2203.12273">DAN: a Segmentation-free Document Attention Network for Handwritten Document Recognition</a>"<br><br></p>
format Recurso digital
id zenodo_https___doi_org_10_5281_zenodo_17185990
institution Zenodo
language fra
publishDate 2025
publisher Zenodo
record_format zenodo
spellingShingle A Document Attention Network (DAN) model dedicated to Information Extraction from 19th century French Land Registry tables
Tual, Solenn
Abadie, Nathalie
Chazalon, Joseph
Duménieu, Bertrand
Perret, Julien
<p>This repository contains the weights of a Document Attention Network (DAN) (Coquenet et al.) model fine-tuned to perform extraction information from digitised tables of the 19th century French Land Registry. </p> <p>The dataset used to train and evaluate the model is available on Zenodo : <a href="https://zenodo.org/records/15411507" rel="nofollow">10.5281/zenodo.15411507</a>. Annotation model is described in the dataset repository. </p> <p>Be aware that this model has been trained with images from the Departement of Valde-Marne. It may not suceed to extract informations from land registry documents from other departements because of other layouts, local languages, etc.</p> <p><a href="https://gitlab.teklia.com/atr/dan">atr-dan</a> Python library has been used to train the model.</p> <p>DAN architecture as been presented in the following paper : "Denis Coquenet, Clément Chatelain, Thierry Paquet, <a href="https://arxiv.org/abs/2203.12273">DAN: a Segmentation-free Document Attention Network for Handwritten Document Recognition</a>"<br><br></p>
title A Document Attention Network (DAN) model dedicated to Information Extraction from 19th century French Land Registry tables
url https://doi.org/10.5281/zenodo.17185990