Back to homepage utwente.nl
Interactional behavior coordination in Human Computer Interaction
Description of research
People include a diverse multitude of culture, differing growing-up background reflected in unique language, customs, social norms and expectations. Sometimes a harmonious communication is difficulty to continue for a long time between two or more persons, especially if they support different opinions or when they are in competition. Socially intelligent devices or systems capable of sensing or detecting information relevant to human social interaction and making it readily available to people—has the potential to improve interactions between different people by taking distinct growing-up background, experience or personal differences into account and offering ways to create greater feelings.
It is considered that embodied behavior coordination will provide important clues to the future investigations of human interactions, mainly in two ways. The first is as an indicator of cooperativeness and empathy, among others. The second is its application as a means to enriching communication. The impact of a practical technology to mediate human interactions in real time would be enormous both for society as a whole (improving business relations, cultural understanding, communication relationship, etc). It would find immediate applications in areas such as adapting interactions to help people with less confidence, training people for improved social interactions or in specific tools for tasks such as negotiation. This technology would also strongly influence science and technology (providing a powerful new class of research tools for social science and anthropology, for example). While the primary goal of such an effort would be to facilitate direct mediated communication between people, advances here would also facilitate interactions between humans and machines.
My research direction falls into the field of Multimodal Human Computer Interaction (MHCI). The goal of this research is to provide more natural and more human-centered HCI systems through studies in Human Behavior Analysis, Computer Vision, and Psychology and so on. Thus, the main focus of my research is on machine learning for modeling multimodal human-machine interactive modes from multi-sensory observations. The related research areas include realizing and analyzing human interactive actions, integrating multiple sensors and pertinent modalities according to the model of the human sensory system. These models will be further used for seamless and proactive design of home, customer market and health support appliances, living and working spaces, and interactive devices including automobiles and robots.
Advisor(s)
Prof. dr. Ir. Anton Nijholt, University of Twente, NL
Prof. dr. ir. Maja Pantic, Imperial Colledge, UK and University of Twente, NL
Dr. Mannes Poel, University of Twente, NL
Duration
1 March 2010 – 1 March 2014
Project
SSPNET: Social Signal Processing Network:
· |
wp8 :Analysis and Interpretation of Social Signals in Political Debates |
· |
wp10: Analysis and synthesis of behavior in face-to-face groups |
Funding institution
SSPNET is a new Network of Excellence funded under work programme topic ICT-2007.2.2 of the European Commission's Seventh Framework Programme. It is coordinated by the IDIAP Research Institute in Switzerland, with funding running from 1 Feb 2009 for 60 months.
Strategic Research Orientation
Natural Interaction in Computer Mediated Environments (NICE)
Links to relevant web pages:
Pictures
