Computational Study of Nonverbal Social Communication

The goal of this project is to develop computational models for predicting human nonverbal behavior during specific social contexts, such as the verbal and nonverbal behaviors immediately preceding a conversational turn. The ability to collect, analyze and ultimately predict human nonverbal cues can provide new insights into human social processes and new human-centric applications that can understand and respond to this natural human communicative channel. They have broad applicability for synthesis of natural animations, multicultural training and the diagnoses of social disorders.