Curriculum
Computer Science for Societal Challenges and Innovation, XXXIX series
Grant sponsor
Università degli Studi di Padova
Supervisor

Alessandro Sperduti
Co-supervisor
s
TBD
Contact
belkoerrolyanis.diallo@studenti.unipd.it

Project description
The project "Deep Learning-Based Architectures for Semantics Discovery of Entities and Events" addresses the challenge of processing and understanding multimedia data—images, videos, texts, and audio. This complexity arises from the gap between multimedia representation and machine perception of entities, their semantic meanings, and interconnected events. The objective is to develop deep learning architectures that link visual entities in images and textual descriptions (derived from audio) to knowledge databases, enabling semantic descriptions through ontology. By semantically integrating and describing entities and events across various data formats, this project aims to enhance a robot's ability to interpret and communicate its understanding of the environment. Advancements in this field will improve multimedia data processing and semantic understanding, facilitating better human-robot interactions and automated multimedia analysis.