Description
Date depot: 1 juillet 2022
Titre: Deep learning approaches and spatial relations reasoning for interpreting Byzantine seals
Directrice de thèse:
Isabelle BLOCH (LIP6)
Encadrante :
Victoria EYHARABIDE (STIH)
Domaine scientifique: Sciences et technologies de l'information et de la communication
Thématique CNRS : Intelligence artificielle
Resumé: The general aim of the thesis is to combine computer vision, knowledge engineering, and mathematical
modeling of spatial relationships to help with the interpretation of Byzantine seals. This research aims to
(i) work on the recognition of objects on seals to analyze iconographic scenes; (ii) estimate the inception
date of Byzantine seals; and (iii) propose solutions based on hybrid AI techniques to interpret damaged
areas based on existing insights. The objective is to explore new artificial intelligence methods applied
to Byzantine sigillography: instance segmentation with deep learning, knowledge graph embeddings, and
spatial reasoning for image understanding. The thesis is expected to contribute to the intersection of
computer vision, knowledge representation, and spatial logic.
Doctorant.e: Sendogan Ege