3D Facial Action Units Recognition for Emotional Expression
This work addresses facial expression analysis for applications like human-computer interaction, but it appears incremental as it applies existing methods (SVM and NN) to AU recognition without major innovation.
The paper tackled the problem of recognizing facial Action Units (AUs) for emotional expression by using facial distances from twelve features, with methods including SVM and Neural Network training. The result involved classification results presented for different SVM kernels and NN phases, but no concrete numbers were provided.
The muscular activities caused the activation of certain AUs for every facial expression at the certain duration of time throughout the facial expression. This paper presents the methods to recognise facial Action Unit (AU) using facial distance of the facial features which activates the muscles. The seven facial action units involved are AU1, AU4, AU6, AU12, AU15, AU17 and AU25 that characterises happy and sad expression. The recognition is performed on each AU according to rules defined based on the distance of each facial points. The facial distances chosen are extracted from twelve facial features. Then the facial distances are trained using Support Vector Machine (SVM) and Neural Network (NN). Classification result using SVM is presented with several different SVM kernels while result using NN is presented for each training, validation and testing phase.