Seminar, part of the conference program of DEAF04.
The seminar aims to give an idea of the state of the art in affective systems by presenting both theoretical and experimental results from different disciplines. A moderated discussion with authors and creators will deal with the assumptions, ambitions and approaches that underpin their work.
Researchers and experimenters in various disciplines are trying to equip machines with some understanding of subtext and context of interaction, rather than only literal commands and feedback. This requires the machine to be able to perceive and reason about emotional cues in human interaction, as in intonation, facial expression and body language, and to form expectations of how its own behavior might affect the human. Conversely, given an adequate internal representation for that purpose, machines could perhaps be made to express emotions in and of themselves. Would such efforts really help improve the quality of human-machine interaction? And could similar techniques be applied in other areas, such as simulation experiments in the social sciences, or synthetic character and narrative generation in the interactive arts?
With Angelika Oei (NL), Rene Verouden (NL), Michelle Teran (CDN/NL), Jeff Mann (CDN/NL), Owen Holland (GB) and Phoebe Sengers (USA). Moderated by Gerd Ruebenstrunk (D).