Category: Workshop

Posted on

Handwritten Text Recognition for the European Digital Treasures Collections.  Hands On workshop by Joan Andreu Sánchez and Enrique Vidal

The first day of  the workshop “New Digital Exponential Technologies Towards The Generation Of Business Models” was concluded by a hands on session led by Joan Andreu Sánchez and Enrique Vidal.

Joan Andreu Sánchez.

Joan Andreu Sánchez is assistant professor at Universitat Politècnica de València and the Director of the Pattern Recognition and Human Language Technologies (PRHLT) Research Center in this university. His main area of research is machine learning and formal languages applied to text recognition and math recognition.

Enrique Vidal is emeritus professor at the same university and former co-leader of PRHLT research center. For many years Dr. Vidal has focussed his research on handwritten document analysis and recognition leading the development of the probabilistic indexing technology.  Joan Andreu and Enrique are founders of tranSkriptorium, an AI spin-off company.

Enrique Vidal.

The contents of a massive volume of digitised handwritten records in archives and libraries all over the world are practically inaccessible, buried beneath thousands of terabytes of high-resolution images. The image textual content could be straightforwardly indexed for plain-text textual access using conventional information retrieval systems if perfect or sufficiently accurate text image transcripts were available.

However, fully automatic transcription results generally lack the level of accuracy that is required for reliable text indexing and search purposes. On the other hand, the massive volume of image collections typically considered for indexing render manual or even computer-assisted transcription as entirely prohibitive. Dr. Sanchez and Dr. Vidal explain how very accurate indexing and search can be directly implemented on the images themselves, without explicitly resorting to image transcripts; they present the results obtained using the proposed techniques on several relevant historical data sets. The results have led to a high interest in these technologies.

You can watch the session on YouTube here and the paper presented at the workshop here: Part I & Part II.

Written by Leonard Callus and the European Digital Treasures Team.

Posted on

ICARUS Convention #28 in Paris with European Digital Treasures workshop

SAVE THE DATE

After two years of online conventions and zoom conferences, we are happy to announce that the upcoming ICARUS Convention #28 will be held in person in Paris from 23rd to 25th of May, 2022 as a hybrid event!

The conference will take place in the conference center of Campus Condorcet in Paris-Aubervilliers and is organised by the Institut de Recherche et d’Histoire des Textes (CNRS) with the support of the French Ministry of Culture and the National Archives (Archives nationales).

Within the programme of the convention, the European Digital Treasures project will hold their workshop “New Business & Conceptual models” led by Yvan Corbat!

One of the key objectives of the Digital Treasures project is to generate a greater added value, profitability, visibility and economic return of European archives, through the identification and implementation of new business models and activities.

The workshop will include practical examples of new activities being implemented by some partners of this project:

The programme of the convention will be finalized within the next days and weeks.
First prospect, further information, details and registration: https://icarus-28.sciencesconf.org/resource/page/id/2

Any questions? Please contact: info@icar-us.eu

More information to come soon – stay tuned!

We are looking forward to seeing you in Paris!

Written by ICARUS & the Digital Treasures team.

Posted on

St. George on a bike: Enrichment of metadata of Cultural Heritage Objects using deep learning. – by Artem Reshetnikov

The European Digital Treasures team continues with the presentations of the experts who participated in the workshop “New Digital Exponential Technologies Towards The Generation Of Business Models” on 2nd and 3rd of September, 2021 at the Provincial Historical Archive of Alicante (Spain). 

Artem Reshetnikov.

The second speech was held by Artem Reshetnikov who is a deep learning researcher at Barcelona Supercomputing Center. Working in several companies and research centers, he received big experience in Computer Vision and Natural Language Processing and applying it to the tasks of different domains. For a long time, he was thinking about how to combine his two main passions: machine learning and art. The solution is the project where he works now: Saint George on a Bike is a project about the enrichment of metadata of paintings using Deep Learning and NLP approaches. 

Abstract.“Saint George on a Bike” project proposes several novel approaches to enrichment of metadata (captions, tags, relationships between objects, iconographic description) for the Cultural Heritage domain, which relies on combining Deep Learning and semantic metadata about paintings. Working with cultural heritage presents challenges not existent for every-day images. Models for objects detection or caption generation are usually trained with datasets that contain correct descriptions of current images or labels for objects, which were generated manually. Apart from this conceptual problem, the paintings are limited in number and represent the same concept in potentially very different styles. Finally, the metadata associated with the images is often poor or inexistent, which makes it hard to properly generate quality metadata. Our approach can assist in generation of metadata for different tasks. By taking into account an exiting metadata of Cultural heritage objects and additional techniques, we can generate tags, relationships between objects or descriptive text which is likely to be directly related to the scene depicted in an image.

You can watch the whole session on YouTube here and read the manuscript paper here!

Written by Artem Reshetnikov & the European Digital Treasures Team.