Dear Colleagues, [With apologies for cross-posting] KBR invites you to attend the KBR Digital Heritage Seminar Series, in cooperation with ULB, UGent, VUB and UCL. In this virtual series running from February to June 2022, three scholars will present their work on image processing in the context of cultural heritage. Tuesday 22 February 2022 at 2 pm CET Thomas Smits, University of Antwerp, “The Visual Digital Turn: Computer Vision and the humanities” https://www.kbr.be/en/agenda/computer-vision-and-the-humanities/ Abstract Digital humanities research has focused primarily on the analysis of texts. However, digital and digitized sources also contain large numbers of images. Using several research projects as examples, I’ll discuss how scholars in the humanities can apply computer vision techniques in their research. Among other things, these new visual AI methods allow researchers to analyze patterns of (re)circulation of images, analyze patterns of media (use), analyze patterns in visual content at scale, analyze patterns of representations (gender/social class), analyze patterns in visual scenes and analyze patterns of visual style. In short, I’ll show how these techniques have opened up the visual side of the digital turn. Tuesday 15th March 2022 at 2 pm CET Apostolos Antonacopoulos, University of Salford, Manchester, UK “Understanding information-rich documents: experiences with historical England & Wales censuses” https://www.kbr.be/en/agenda/dhs-censuses/ Abstract In the Pattern Recognition and Image Analysis (PRImA) research Lab, from the University of Salford, UK, great efforts have been devoted to the investigation and analysis of cultural heritage materials especially using image processing and pattern recognition techniques. In the last a few years, a series of large-scale projects, including IMPACT - Improving Access to Text, Europeana Newspapers, Succeed - Support Action Centre of Competence in Digitisation, EMPO - Early Modern OCR project and Census 1961 were being hosted by PRImA. In this talk, Prof. Antonacopoulos, the founder and leading professor of PRImA, will give his insights on the processing and exploitation of cultural heritage materials. A special focus will be given to the ongoing project of Census-1961, where the main objective is to extract and model historical population census information using computer vision techniques. Monday 11th April 2022 at 2 pm CET Clemens Neudecker, Berlin State Library, Berlin, Germany “New Tools for Old Documents – Layout Analysis and OCR with Deep Learning and Heuristics” https://www.kbr.be/en/agenda/dhs-layout-analysis/ Abstract This talk will discuss the main achievements and experiences of the QURATOR project at the Berlin State Library (SBB) for document layout analysis. Historical documents that are being digitized in large quantities by libraries and archives frequently exhibit a wide array of features that disturb layout analysis, such as complex layouts with multiple columns, drop capitals and illustrations, skewed or curved text lines, noise, annotations, etc. In order to deal with these challenges and defects, a robust document layout analysis was developed that is implemented by pixel-wise segmentation using convolutional neural networks. In addition, heuristic methods are applied to detect columns or marginalia, and to determine the reading order of text regions. A key objective lies in feeding the resulting outputs to subsequent processes like a text recognition (OCR) engine or an image similarity search. Practical information: These talks will be held, via Microsoft Teams, in English, with questions in French, Dutch or English. The target audience is scholars, but the general public is warmly welcome. Registration is free but mandatory: https://forms.gle/v4Cz4jgdQ1yb33zY6. You will be sent the link to the virtual meeting on the morning of the event. This series is co-organised by KBR’s growing research labs which includes: the Digital Research Lab, Data Science Lab, and the Digital Literary Lab in cooperation with Université libre de Bruxelles, Ghent University, Vrije Universiteit Brussel, and Université catholique de Louvain. The KBR DH Series Team Julie Birkholz, Sébastien de Valeriola, Tan Lu & Isabelle Gribomont