Virtual Agents and Anthropomorphism: Expressivity, Emotions, Opinions, and Beyond

Description

Affective computing is the study and development of systems (video game agents or robots) that have the capacity to recognise, express and influence human affect, often based on signal processing and modelling approaches for verbal and non-verbal signals: from voice, face or body. This research will be focused on deep learning approaches for affective computing and signal processing, and on studying several factors of variability in expressivity: emotions, but also opinions and personality, related to longer terms character traits that people could assign to virtual agent or robots through anthropomorphism. The researcher will be integrated within a significant group of deep learning researchers involved in regional and international collaboration with worldleading labs in the area.