For most of us, social perception is used unconsciously for some of the most important actions we take in our life: negotiating economic and affective resources, making new friends, establishing credibility, or leadership. Social Signal Processing and Affective Computing are emergent areas of research that focus on the analysis of social cues and personal traits.

The basic signals come from different sources and include gestures, such as scratching, head nods, “huh” utterances or facial expressions. As such, automatic systems in this line of work benefit of technologies such as face detection and localization, head and face tracking, facial expression analysis, body detection and tracking, visual analysis of body gestures, posture recognition, activity recognition, estimation of audio features such as pitch, intensity, and speech rate, and the recognition of non-linguistic vocalizations like laughs, cries, sighs, and coughs.

Publications

S. Escalera, R. M. Martínez, J. Vitrià, P. Radeva, M. T.  Anguera. Detección automática de la dominancia en conversaciones diádicas. Escritos de Psicología, Vol. 3, nº 2, pp. 41-45, 2010.

M.Rojas, D.Masip, A.Todorov, J.Vitrià, Automatic Point-based Facial Trait Judgments Evaluation, IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2010.

S. Escalera, O. Pujol, P. Radeva, J. Vitrià and M. Teresa Anguera.Automatic Detection of Dominance and Expected Interest. EURASIP Journal on Advances in Signal Processing, 2010.

B. Raducanu, J.Vitria, and D. Gatica-Perez. You are Fired! Nonverbal Role Analysis in Competitive Meetings. In Proc. IEEE Int. Conf. on Acoustics, Speech, and Signal Processing (ICASSP), Taiwan, Apr. 2009.

Advertisements