Digital Palaeography: New Machines and Old Texts (Dagstuhl, 20-24 July 2014)

The very successful and fruitful Dagstuhl seminar 14302, jointly organized by the Oriflamms project leader, gathered 24 Humanists and Computer Scientists from 10 countries. The report will soon appear. In the meanwhile, here is the abstract of the ORIFLAMMS’ presentation.

Digital Palaeography. Text-Image Alignment and Script/Scribal Variability (ANR ORIFLAMMS / Cap Digital)

Medieval scripts are a challenge to historical analysis, as for describing and representing the graphical evidence, analyzing and clustering letter forms and their features through Computer Vision and analyzing historical phenomena. The ANR funded research project ORIFLAMMS (Ontology Research, Image Feature, Letterform Analysis on Multilingual Medieval Scripts, 2013-2016) gathers seven partners from the Humanities and Computer Science (IRHT = Institut de Recherche et d’Histoire des Textes, CNRS; CESCM = Centre d’Études Supérieures de Civilisation Médiévale; École Nationale des Chartes; ICAR = Interactions Corpus Apprentissages Représentations, École Normale Supérieure, for the Humanities; A2iA, LIRIS = Laboratoire d’InfoRmatique en Image et Systèmes d’information, INSA Lyon; LIPADE = Laboratoire d’Informatique de Paris Descartes, for Computer Science). It aims at studying the coherence and variability of graphical systems, according to their language, level of formality, support, genre, date and place, as well as creating an ontology of medieval signs, through the alignment of text and images, by extracting letterforms, abbreviations and signs, then perform pattern similarity analysis, and enhance the results with computational linguistics and paleographical analysis. In order to achieve representative results, several core corpuses have been identified (charters, books, books of charters such as cartularies and registers, inscriptions). The research is based on XML-TEI compliant editions and compels to deepening our understanding of scribal systems and forms [1, 2]. As part of this research, a software has been developed in order to easily visualize and validate the text-image alignment. The latter is produced by two different systems developed in this project: the first one without prior knowledge [3], the second one with GMM and DNN with very good results. By now, two large data sets have been aligned: Queste du Graal including 130 pages, 10700 lines, more than 115’000 words and 400’300 characters; Fontenay including 104 pages, 1341 lines, more than 22’200 words and 99’900 characters. This is a major first step. With the following corpuses, this research contributes to both Humanities (letterform identification, historical semiotics) and Computer Science (Handwriting recognition), with the core idea of not reinventing the wheel, but using former research, computer and human brain at their maximal capacities.


The research leading to these results has received funding from the Agence Nationale de la Recherche and Cap Digital under grant agreement no. ANR-12-CORP-0010.


1 D. Stutzmann. Paléographie statistique pour décrire, identifier, dater … Normaliser pour coopèrer et aller plus loin? In Kodikologie und Paläographie im digitalen Zeitalter 2 – Codicology and Palaeography in the Digital Age 2, Norderstedt, 2010, pp. 247–277

2 D. Stutzmann. Ontologie des formes et encodage des textes manuscrits médiévaux. Le projet ORIFLAMMS, In Document numérique, 16/3 (2013):81–95. DOI: 10.3166/DN.16.3.69-79.

3 Y. Leydier, V. Eglin, S. Bres, D. Stutzmann. Learning-free text-image alignment for medieval manuscripts. In Proc. Int. Conf. on Frontiers in Handwriting Recognition, Crete, Greece, 2014