Thesis Subject: Deep learning-based curation and enrichment of digital heritage
Director: Prof. Abdelaziz Bouras
Co-Director: Prof. Sebti Foufou (Univeristé de Bourgogne)
Abstract: Cultural heritage (CH) takes an important part in defining the identity and the history of a civilization or a nation. Physically preserving CH assets for the long term is not effective and may induce multiple risks related to destruction and damage. Digital technologies such as photography and 3D scanning provided new alternatives for digital preservation. However, adapting them for CH is a challenging task. In fact, fully digitizing cultural assets (visually and historically) is only easy when it comes to assets that are rich physically (in a good shape) and all their data is at possession (fully annotated). However, in the real world, many assets suffer from physical degradation and information loss. Usually, to annotate and curate these assets, heritage institutions refer to art specialists, historians and other institutions. This process is tedious, involves many time and financial resources and can often be inaccurate. Through this thesis, we aim at studying new techniques that improve the process of cultural data digitization. Our challenges are mostly related to the asset annotation and enrichment process which is ineffective manually and suffer from missing and incomplete data problems. We introduce various techniques and approaches that rely on advanced machine learning and deep learning tools to tackle these challenges that consists in automatically completing missing cultural data. We mostly focus on two major types of missing data: missing textual data (metadata) and missing visual data (missing patches or parts).