Micro-Expression Spotting Using Facial Landmarks

Abstract

Micro-expressions (MEs) are brief involuntary facial expressions and it usually happens when people try to mask their emotions in high-stake situations. As a result, it is hard to detect the occurrence of a spontaneous MEs due to the limitation of human vision in spotting the brief and subtle change of facial expression. A MEs recognition system can be categorized into two major tasks which are the spotting of the fleeting change of facial expression and the classification of the emotion behind the spotted MEs. In fact, most of the spotting of MEs proposed in the earlier work by other researchers have low accuracy and complex system model. As a result, this paper proposed a MEs spotting system that requires no training and it spots the spontaneous MEs from the videos by detecting the changes in the ratio of the Euclidean distances of facial landmark in three facial regions (left eyebrow, right eyebrow and mouth) .The proposed method was evaluated on CASME II dataset with an average accuracy of 64.77 % and the highest accuracy was 82.30 %.