Indonesia Conference Directory

Micro-expression recognition is one of the popular researches in analysing expressions on the face. Micro-expression is a facial movement that occurs in a short time and is difficult to identify manually by human eyes. In general research, facial landmarks are used to form a large size ROI for each facial feature for the feature extraction process. In this study, we utilize certain points in the face to trace the subtle motion of the face for the introduction of micro-expressions in the sequence of onset images to offset. The method of determining the landmark points to be compared is the Discriminative Response Map Fitting method and the Active Shape Model method. This research proposes to use 19 points in the area of the eyebrows, eyes, and mouth. To measure the subtle motion tracking of facial features in each frame tracking is done using the Kanade Lucas Tomasi method. The features obtained are the differences in coordinate values between frames, magnitude, and theta. We evaluated the method proposed in CASME2 and SAMM showed that using naive bayes is 74.6% and neural network is 63.5% classification accuracy in SAMM. In CASME2 using naive bayes is 79.6% and neural network is 61.1% classification accuracy.