Karolina Nurzynska and Bogdan Smolka, Silesian University of Technology, Poland

When automatic recognition of emotion became feasible, novel challenges has evolved. One of them is the recognition whether a presented emotion is genuine or not. In this work, a fully automated system for differentiation between spontaneous and posed smiles is presented. This solution exploits information derived from landmark points, which track the movement of fiducial elements of face. Additionally, the smile intensity computed with SNiP (Smile-Neutral intensity Predictor) system is exploited to deliver additional descriptive data. The performed experiments revealed that when an image sequence describes all phases of smile, the landmark points based approach achieves almost 80% accuracy, but when only onset is exploited, additional support from visual cues is necessary to obtain comparable outcomes.

Smile veracity recognition, Landmarks, Smile intensity, Classification