Research Assistant Intern 0925
January 08, 2009
Project Name
Understanding the role of facial expressions in face-to-face negotiations
Project Description
This collaborative project between ICT’s computational emotion group and USC’s Marshall School of Business examines the relationship between nonverbal behavior (especially facial expressions) and outcomes in face-to-face negotiations. The project involves the use (and possible extension) of a variety of human behavior recognition methods to assess the validity of social psychological theories concerning the social impact of nonverbal signals. This research will feed into the development of advanced virtual humans that are able to simulate realistic emotional displays, teach nonverbal skills, and potentially influence the outcome of negotiations with human users.
Job Description
The intern is expected to play a central role in using behavior recognition techniques to extract dyadic patterns of nonverbal behavior from a corpus of face-to-face negotiations. These patterns should give insight into the impact of nonverbal factors on negotiation outcomes and will be used to drive the behavior of virtual negotiation partners. Recognition techniques include a FACS-based facial expression recognition, head and gaze tracking methods, a variety of speech processing tools, and some physiological measures. The intern will have the opportunity to suggest improvements or alternatives to these methods and explore and/or develop machine learning techniques for extracting patterns from multimodal dyadic datasets.
The ideal candidate will be a PhD student in computer science and have familiarity with:
• Vision-based methods for recognizing and understanding human nonverbal behavior
• Machine learning methods for identifying patterns in time-series data
• Social psychological theories of interpersonal interaction
• Good programming skills in C++/C#, matlab, scripting