Person-Specific Joy Expression Synthesis with Geometric Method

Smiling has a psychiatric effect in emotional state and may hold tremendous potential for clinical remediation in psychiatric disorders. A few researchers in image synthesis work on acting on the emotional state of subjects by automatically deforming their faces to synthesize joyful expression. However, to generate these expressions they apply the same deformation for the subjects while each person smiles differently. In this paper, we head towards a personalized synthesis of the joy expression. We have studied the trajectories of the face landmarks during a smile on the CK, Oulu-CASIA and MMI databases. The obtained results show that the smile is personal, straight and it occurs in a different way for each person. That is why we propose a system that can photo-realistically transform in real-time a detected face into a personalized joyful one using a geometric method. Both visual fidelity and a statistical study demonstrate that our person-specific method can generate personalized joy expressions closer to the ground truth than two non-personal state-of-the-art approaches.