Geometrical Pattern Discovery

How can we detect common structural patterns in large collections of paintings, sculptures, buildings?

How can we classify temporal curves?

How can we detect influences and trace propagation of elements?



Page Layout Analysis of Text-heavy Historical Documents: a Comparison of Textual and Visual Approaches

S. Najem-Meyer; M. Romanello 

Page layout analysis is a fundamental step in document processing which enables to segment a page into regions of interest. With highly complex layouts and mixed scripts, scholarly commentaries are text-heavy documents which remain challenging for state-of-the-art models. Their layout considerably varies across editions and their most important regions are mainly defined by semantic rather than graphical characteristics such as position or appearance. This setting calls for a comparison between textual, visual and hybrid approaches. We therefore assess the performances of two transformers (LayoutLMv3 and RoBERTa) and an objection-detection network (YOLOv5). If results show a clear advantage in favor of the latter, we also list several caveats to this finding. In addition to our experiments, we release a dataset of ca. 300 annotated pages sampled from 19th century commentaries.

Proceedings of the Computational Humanities Research Conference 2022 Antwerp, Belgium, December 12-14, 2022


Third Conference on Computational Humanities Research (CHR 2022), Antwerp, Belgium, December 12-14, 2022.

p. 36-54

DOI : 10.48550/arXiv.2212.13924

Visual Patterns Discovery in Large Databases of Paintings

I. di Lenardo; B. L. A. Seguin; F. Kaplan 

The digitization of large databases of works of arts photographs opens new avenue for research in art history. For instance, collecting and analyzing painting representations beyond the relatively small number of commonly accessible works was previously extremely challenging. In the coming years,researchers are likely to have an easier access not only to representations of paintings from museums archives but also from private collections, fine arts auction houses, art historian However, the access to large online database is in itself not sufficient. There is a need for efficient search engines, capable of searching painting representations not only on the basis of textual metadata but also directly through visual queries. In this paper we explore how convolutional neural network descriptors can be used in combination with algebraic queries to express powerful search queries in the context of art history research.


Digital Humanities 2016, Krakow, Polland, July 11-16, 2016.

Carlo Helman : merchant, patron and collector in the Antwerp – Venice migrant network

I. di Lenardo 

This contribution is part of the monographic number of the Nederlands Yearbook for History of Art dedicated to a large overview on the “Art and Migration. Nethelandish Artists on the Move, 1400-1750”. In the dynamics of migration, circulation, establishing trough Europe in the Modern Era, the network’s analysis play a fundamental role. The essay explores the prominent role played by Antwerp merchants in Venice in forging contacts between artists, patrons and agent of art in promoting the exchange of goods and ideas within their adopted home. In the course of the 16th century, and more particularly towards the end of that period, the complex network of Netherlandish merchant families, operating on a European level, played a crucial role in the circulation of artists, paintings and other artworks in Italy and beyond. The article proposed here deals with Carlo Helman, a Venetian resident of Antwerp origins, a major figure whose importance in this context has been insufficiently studied. Helman’s family firm traded in practically every kind of commodity, ranging from wool and spices to pearls and diamonds, and, indeed, artworks, “in omnibus mundis regnis”, as we read in the commemorative inscription on his monumental tomb in the Venetian church of Santa Maria Formosa. A high-class international trader in Venice, Helman was consul of the “Nattione Fiamenga”. Helman had a conspicuous collection of art, including classics of the “Venetian maniera” like Titian, Veronese and Bassano, but also important pictures by Northern masters. Moreover, his collection contained a remarkable cartographic section. In Venice, Helman had contacts with the Bassano dynasty, Paolo Fiammingo, Dirck de Vries, Lodewijck Toeput (Pozzoserrato) and the Sadeler brothers, artists who, in one way or another, introduced novel themes and typologies on the Italian, and, indeed, European market. The dedication to Helman on a print by Raphael Sadeler, reproducing Bassano’s Parable of the Sower, photographs the merchant’s role in the diffusion of Bassanesque themes in the North. Helman’s connections with the Zanfort brothers, dealers in tapestries and commercial agents of Hieronymus Cock are further indications of the merchant’s exemplary role of collector, merchant and agent of artists in a European network of “art” commerce.

Art and Migration. Netherlandish Artists on the Move, 1400-1750; Leiden: Brill, 2014. p. 325-347.

ISBN : 9789004270534

“Cities of Fire”. Iconography, Fortune and the Circulation of Fire Paintings in Flanders and Italy in the XVI Century.

I. di Lenardo 

The Wounded City. The representation of Urban Disasters in European Art (XV-XX Centuries); Leiden: Brill, 2014. p. 100–115.

ISBN : 9789004284913

DOI : 10.1163/9789004300682_006

Semi-Automatic Transcription Tool for Ancient Manuscripts

M. M. J-A. Simeoni 

In this work, we investigate various techniques from the fields of shape analysis and image processing in order to construct a semi-automatic transcription tool for ancient manuscripts. First, we design a shape matching procedure using shape contexts, introduced in [1], and exploit this procedure to compute different distances between two arbitrary shapes/words. Then, we use Fischer discrimination to combine these distances in a single similarity measure and use it to naturally represent the words on a similarity graph. Finally, we investigate an unsupervised clustering analysis on this graph to create groups of semantically similar words and propose an uncertainty measure associated with the attribution of one word to a group. The clusters together with the uncertainty measure form the core of the semi-automatic transcription tool, that we test on a dataset of 42 words. The average classification accuracy achieved with this technique on this dataset is of 86%, which is quiet satisfying. This tool allows to reduce the actual number of words we need to type to transcript a document of 70%.

IC Research Day 2014: Challenges in Big Data, SwissTech Convention Center, Lausanne, Switzerland, June 12, 2014.