Real Emotion Recognition by Detecting Symmetry Patterns with Dihedral Group
Mehdi Ghayoumi, Arvind Kumar Bansal · 2016
In this paper, we describe how to use the dihedral group theory to detect symmetry on Facial Key Points (FKP) to recognize real emotions. We present an algorithm for symmetry detection with the dihedral group and deploy it for finding symmetry (rotation and reflection) on FKP. The method can be applied in many areas which have the same data texture. The method used for real emotion recognition and experimental results shows that using the dihedral group concept can reach it better in the subject area which contains symmetry in nature because of its robustness.