Bringing the Affective Body Outside the Controlled World: a Challenge for Deep Learning
Affective-aware technologies are becoming popular but they remain confined to very controlled situations using a limited set of channels. These restrictions are due in part to the complexity of the affective phenomena themselves and in part to the current lack of embedding of wearable sensors. Progress is being made regarding the latter with sensors now being integrated in clothes, making it possible to capture with greater bandwidth the contexts in which emotional experiences are taking place. Being able to infer affective states in ubiquitous and uncontrolled scenarios requires new modelling paradigms capable of dealing with high-dimensional unlabelled data. In my talk, I will present what we have learned from our studies on how body and touch explain how we feel and how there is a need and opportunity to extend these studies into our everyday life.
Nadia Bianchi-Berthouze is a Full Professor in Affective Computing and Interaction at the Interaction Centre of the University College London (UCL). Her research focuses on designing technology that can sense the affective state of its users and use that information to tailor the interaction process. She has pioneered the field of Affective Computing, first investigating body movement and more recently touch behaviour as means to recognize, measure and steer the quality of the user experience in full-body computer games, physical rehabilitation and textile design.