Thibault Clérice’s research while affiliated with Centre Jean Perrin and other places

We present a novel, open-access dataset designed for semantic layout analysis, built to support document recreation workflows through mapping with the Text Encoding Initiative (TEI) standard. This dataset includes 7,254 annotated pages spanning a large temporal range (1600-2024) of digitised and born-digital materials across diverse document types (magazines, papers from sciences and humanities, PhD theses, monographs, plays, administrative reports, etc.) sorted into modular subsets. By incorporating content from different periods and genres, it addresses varying layout complexities and historical changes in document structure. The modular design allows domain-specific configurations. We evaluate object detection models on this dataset, examining the impact of input size and subset-based training. Results show that a 1280-pixel input size for YOLO is optimal and that training on subsets generally benefits from incorporating them into a generic model rather than fine-tuning pre-trained weights.

Whether or not several Creole languages which developed during the early modern period can be considered genetic descendants of European languages has been the subject of intense debate. This is in large part due to the absence of evidence of intermediate forms. This work introduces a new open corpus, the Moly\'e corpus, which combines stereotypical representations of three kinds of language variation in Europe with early attestations of French-based Creole languages across a period of 400 years. It is intended to facilitate future research on the continuity between contact situations in Europe and Creolophone (former) colonies.

Given enough data of sufficient quality, HTR systems can achieve high accuracy, regardless of language, script or medium. Despite growing pooling of datasets, the question of the required quantity of training material still remains crucial for the transfer of models to out-of-domain documents, or the recognition of new scripts and under-resourced character classes. We propose a new data augmentation strategy, using generative adversarial networks (GAN). Inspired by synthetic lines generation for printed documents, our objective is to generate handwritten lines in order to massively produce data for a given style or under-resourced character class. Our approach, based on a variant of ScrabbleGAN, demonstrates the feasibility for various scripts, either in the presence of a high number and variety of abbreviations (Latin) and spellings or letter forms (Medieval French), in a situation of data scarcity (Armenian), or in the instance of a very cursive script (Arabic Maghribi). We then study the impact of synthetic line generation on HTR, by evaluating the gain for out-of-domain documents and under-resourced classes.

Layout Analysis (the identification of zones and their classification) is the first step along line segmentation in Optical Character Recognition and similar tasks. The ability of identifying main body of text from marginal text or running titles makes the difference between extracting the work full text of a digitized book and noisy outputs. We show that most segmenters focus on pixel classification and that polygonization of this output has not been used as a target for the latest competition on historical document (ICDAR 2017 and onwards), despite being the focus in the early 2010s. We propose to shift, for efficiency, the task from a pixel classification-based polygonization to an object detection using isothetic rectangles. We compare the output of Kraken and YOLOv5 in terms of segmentation and show that the later severely outperforms the first on small datasets (1110 samples and below). We release two datasets for training and evaluation on historical documents as well as a new package, YALTAi, which injects YOLOv5 in the segmentation pipeline of Kraken 4.1.

Machine learning begins with machine teaching: in the following paper, we present the data that we have prepared to kick-start the training of reliable OCR models for 17th century prints written in French. The construction of a representative corpus is a major challenge: we need to gather documents from different decades and of different genres to cover as many sizes, weights and styles as possible. Historical prints containing glyphs and typefaces that have now disappeared, transcription is a complex act, for which we present guidelines. Finally, we provide preliminary results based on these training data and experiments to improve them. L'apprentissage machine commence avec l'enseignement machine : dans cet article, nous présentons les données que nous avons préparées pour entraîner des modèles OCR fiables pour les imprimés du XVIIe siècle écrits en français. La construction d'un corpus représentatif est un enjeu majeur : il faut rassembler des documents de différentes décennies et de différents genres pour couvrir un maximum de tailles, de graisse et de styles. Les imprimés historiques contenant des glyphes et des caractères aujourd'hui disparus, la transcription est un acte complexe, pour lequel nous présentons des lignes directrices. Enfin, nous fournissons des résultats préliminaires basés sur ces données d'entraînement et des expériences pour les améliorer.

This paper presents a novel segmentation and handwritten text recognition dataset for Medieval Latin from the 11th to the 16th century. It connects with Medieval French datasets, as well as earlier Latin datasets, by enforcing common guidelines, bringing 263,000 new characters and now totaling over a million characters for medieval manuscripts in both languages. We provide our own addition to Ariane Pinche’s Old French guidelines to deal with specific Latin cases. We also offer an overview of how we addressed this dataset compilation through the use of pre-existing resources. With a higher abbreviation ratio and a better representation of abbreviating marks, we offer new models that outperform the Old French base model on Latin datasets, improving accuracy by 5% on unknown Latin manuscripts.

A lot of available digitized manuscripts online are actually digitized microfilms, a technology dating back from the 1930s. With the progress of artificial colorization, we make the hypothesis that microfilms could be colored with these recent technologies, testing InstColorization. We train a model over an ad-hoc dataset of 18 788 color images that are artificially gray-scaled for this purpose. With promising results in terms of colorization but clear limitations due to the difference between artificially grayscaled images and "naturaly" greyscaled microfilms, we evaluate the impact of this artificial colorization on two downstream tasks using Kraken: layout analysis and text recognition. Unfortunately, the results show little to no improvements which limits the interest of artificial colorization on manuscripts in the computer vision domain.