CLIENT:

University of Rostock

PROJECT:

Back-of-the-book indices of cited passages

We worked with the Epische Bauformen team to build a semi-automatic workflow for the creation of back-of-the-book indices of cited passages for the compendium "Structures of Epic Poetry", a publication in four volumes edited by C. Reitz and S. Finkmann, and published by De Gruyter (2019). The resulting 4 indices list more than 15,000 citations, automatically extracted from over 2,700 pages and carefully double checked by the book editors and their collaborators. We also developed a multi-index web application that gives the book readers a powerful way of browsing and searching through the publication contents.

CLIENT:

The IOC Olympic Studies Centre / Olympic Foundation for Culture and Heritage

PROJECT:

Olympic Knowledge

Platform

Odoma is closely collaborating with The Olympic Studies Centre on the conceptualisation of an Olympic knowledge platform, on which users should be able to search, engage, create and share Olympic knowledge. 
The development of this project is expected to start in the course of 2020.

CLIENT:

Watches Manufacture 

PROJECT:

Deep Reading on handwritten characters

Extraction and transcription of 2M handwritten records that were searched manually on a daily basis.

By using our own tool "Groundtruth Manager" to feed the Machine Learning algorithm developed specifically for our client, we delivered more than 90% of records with 99% confidence score.

CLIENT:

Odoma - product

PROJECT:

Scholar Map

The Scholar Map is a mobile friendly interface to explore personal collections of academic literature and share it with other. By using Topic Modeling to visualize and map the content, the user can quickly browse through thousands of documents and learn the key elements of one topic.

CLIENT:

Cité des Sciences - Paris

PROJECT:

Image similarities Explorer

Developed by Odoma for the Terra Data Exhibition in Paris, in partnership with EPFL/DHLab, this is an interface to navigate through 30’000 pictures using only structural analysis of the images. Aided by a Neural Network algorithm, users could simply select one image to display all similar images.

CLIENT:

Fondation Vallotton + ISEA

PROJECT:

Félix Vallotton - Illustrateur

The Vallotton foundation and the Swiss Institute for Arts studies asked Odoma to conceptualize, from a technical and user experience point of view, an open access platform for Félix Vallotton catalogue raisonné. A collection of several thousands images are interconnected through a system capable of showing the circulation of patterns from one image to the other.

CLIENT:

HEAD Geneva + FNS

PROJECT:

Extra-terrestrial architecture 

We are currently working with Professor Christophe Kihm (HEAD Geneva), on a 4 years research project about space architecture. We are building a database and a dedicated search engine to access a large collection of multimedia data. We are developing and experimenting new ways of searching, comparing and circulating through the vast complexity of the space iconography. 
Contact
Odoma Sàrl
Avenue Dickens 2
1006 Lausanne
Switzerland
0041 79 211 45 69