Jump to: Welcome | Seminar on Collaborative Annotation and IIIF | About the Mezanno Project
Mezanno: Collaborative and Assisted Annotation of Public Textual Resources Published with the IIIF API
Welcome to the public website for the Mezanno project (Fall 2024 — Summer 2026).
Our goal is to facilitate the collaborative annotation of textual resources published with the IIIF API, primarily through enhancements to the Mirador viewer. This project is a collaboration between the National Library of France (BnF), the EPITA Graduate School of Computer Science, the French National Institute of Geographic and Forest Information (IGN), and the School for Advanced Studies in the Social Sciences (EHESS).
The project is currently in its early stages. Our first action consists in organizing a seminar on “Collaboratif Annotation and IIIF” at the French National Library (BnF), on Jan. 24, 2025. It’s preliminary program is available below.
— Mezanno Team: N. Abadie (IGN), E. Carlinet (EPITA), J. Chazalon (EPITA), P. Cristofoli (EHESS), B. Dumenieu (EHESS), J.-P. Moreux (BnF), J. Perret (IGN)
📝 Registration is free but mandatory: https://bit.ly/20250124-seminaire-mzn-bnf
🔵 Presentation of the Mezzano Project
🔵 Presentation of the Glycerine Platform by Ian Mc Crabb, Director of Systemik Solutions (Australia)
Glycerine offers a suite of annotation and workflow tools enabling researchers, curators, and students to collaborate on Humanities and Social Sciences (HSS) projects. Images can be organized into collections and shared with collaborators. Annotation sets combine semantic tags from domain-specific vocabularies with critical analysis in multiple languages. Annotated images can be published as research outputs in the form of visualizations and archived in durable formats.
🔵 Presentation of the PubLink Project by Elisa Bastianello (Bibliotheca Hertziana)
PubLink is developing a testbed for enriching journal articles and academic documents, designed as a companion to the OJS infrastructure. Three types of enrichment are distinguished: Linking entities such as places, people, and artworks. Connecting cited literature to library catalogs, enabling clear referencing of works without a DOI. Integrating IIIF images and annotations via a built-in annotation server. These enriched texts are better indexed by search engines and enhance the visibility of interdisciplinary connections in scientific disciplines.
☕ Coffee Break 🥐
🔵 Presentation of the Adno Platform by Thierry Pasquier, Project Manager for “Digital Cultures,” Espace Mendès France, Center for Scientific, Technical, and Industrial Culture, Poitiers, Nouvelle-Aquitaine
Adno is a free, user-friendly web application for creating annotation series on IIIF-compliant images and beyond. Designed in collaboration with cultural and scientific mediators, it allows the creation, presentation, and sharing of guided or self-guided tours within images. It is also of interest to researchers and has potential applications for a broad audience.
🔵 Presentation of the Arkindex Platform by Christopher Kermorvant, President of TEKLIA, Specialist in Digitized Document Processing with AI
Arkindex is an open-source platform for digitized document processing developed by TEKLIA since 2019. The goal of Arkindex is to handle any type of document, using any algorithm, at any scale.
🔵 Presentation of the eScriptorium Platform by Peter Stokes, Director of Studies at EPHE-PSL and Co-Director of eScriptorium
eScriptorium is a platform for the automatic transcription of handwritten and printed documents (HTR). The software is free, open-source, and designed to handle a wide variety of scripts.
🔵 Presentation of the AIKON Platform by Ségolène Albouy, Research Engineer of the Imagine Team at LIGM, Member of the ERC Discover Project
AIKON is a modular platform for humanities researchers, combining artificial intelligence and computer vision to analyze large-scale heritage collections. It offers tools for visualizing, extracting, and analyzing historical illustrations while supporting interdisciplinary collaboration and the sustainability of digital projects.
🔵 Presentation of the Arvest Platform by Jacob Hart, Post-doctoral fellow at the Université Rennes 2 and member of the project From Stage to Data
Arvest is an open-source, IIIF-driven platform designed for multimodal research, enabling the creation, annotation, and navigation of document networks across images, videos, audio, and texts. With machine learning capabilities and an accessible API, Arvest integrates seamlessly into complex data processing workflows. It facilitates fluid analysis, supporting both close and distant perspectives while maintaining a strong connection to primary sources.
Recent technological breakthroughs in fields such as artificial intelligence, the creation of cross-platform user interfaces, and the ease of deploying services have opened new avenues for research in social sciences. Large corpora of serial historical sources, such as censuses, directories, dictionaries, cadastres, or official publications, can now be processed semi-automatically to generate fine-grained, large-scale quantitative data of sufficient quality for research purposes. However, while the associated scientific methodology is becoming more refined, researchers still lack tools to efficiently query, organize, and construct their research objects from digitized archival collections. This challenge is even greater when dealing with massive corpora that are impossible to exploit without appropriate tools.
The Mezanno Project, organizer of this workshop, aims to create a suite of open and interoperable tools for assisted annotation of customized corpora, supported by a community of users and contributors. These tools are designed to address three key stages:
Mezanno will rely heavily on the IIIF standard to easily assemble corpora from public resources and will integrate artificial intelligence modules to assist users in extracting or transcribing raw content from the documents.