Generation of Facial Expressions through Manipulating Facial Parameters

Abstract

Extracting  discriminative features from salient facial patches plays an important role in effective facial expression recognition. The accurate detection of facial landmark needed to improve the localization of the salient patches on face images. This paper proposes framework for expression recognition by using appearance features of  facial patches. By synthesize of emotional facial expressions through manipulating facial parameters, we aim to introduce facial emotion to a neutral facial image. For this, we perform a preprocessing process in which the face region is identified based on the skin color segmentation method. Then face is located with the connected area identification process. The feature points in facial region isexamined, extracted and get stored. In this algorithm bilinear transformation is used with in the mesh warping technique to manipulate the facial parameters. Thus we could generate some standard set of expressions like happy, angry, sad, fear, surprise.However, the optimal implementation of the proposed framework will significantly improve the computational cost and real-time expression recognition can be achieved with substantial accuracy. Further analysis and efforts are required to improve the performance by addressing some issues.

Cite this Article
Navami VS, Sindhiya. MP. Generation of Facial Expressions through Manipulating Facial Parameters. Journal of Artificial Intelligence Research & Advances. 2016; 3(2): 33–38p.