I was involved in the EU FP7 Tardis Project. TARDIS aimed to build a scenario-based serious-game simulation platform for young people (ages 18-25) at risk of exclusion to explore, practice and improve their social skills. TARDIS facilitates the interaction through virtual agents (VAs) acting as recruiters in job interviews scenarios. My contribution in this project focused on the virtual recruiter. I proposed a data-driven model allowing VAs to express different interpersonal attitudes (dominant, friendly…) which takes into account how the recruiter’s non-verbal behaviors are sequenced. The model uses a dataset of non-verbal behavior sequences obtained using sequence mining techniques, and can generate new sequences to express a chosen interpersonal attitude.

Mathieu Chollet
Lecturer - Virtual Social Interactions

My research interests include interactive virtual humans and multimodal behaviour understanding through social signal processing. My application domains include social skills training, healthcare and wellbeing, and team collaboration.