Output details
11 - Computer Science and Informatics
Queen Mary University of London
Affect recognition for interactive companions: Challenges and design in real world scenarios
<23>The paper reports a framework and results for a real time affect recognition and decision systems capable of sustaining an meaningful, empathic relationship over an extended period of time in a real social situation. Unlike most previous work this system functions fully autonomously and in a children's chess club in Lisbon. The system has to cope with the subtle differences in children's expressions over those in e.g. adult posed datasets. The working system was also demonstrated at the 2012 Cbit industry trade fair and elements presented at Robotville at the London Science Museum in 2012 to over 4000 http://www.sciencemuseum.org.uk/robotville.