Facial Skin Smoothening by Iterative Steering Kernel Regression Technique

Abstract: The paper enhanced the facial skin using iterative steering kernel regression which is based on edge preserving energy minimization. The region-aware mask allows the user to beautify faces with three major skins attributes homogeneity, lighting, and color. The main aim of this paper is to beautify the skin by decomposition into the three layers using edge preserving filter. After layer decomposition a facial landmarks and significant features extracted from the database to generate the mask into three different layers which allow the user to perform facial beautification simply by adjusting its skin parameter. The combination of these parameters can be optimized automatically it depend on the data base or knowledge base. The experiment result shows that the image is enhanced more as compare to the original image. After three region-aware masks have been obtained, a user can perform facial beautification simply by adjusting the skin parameters. Result of proposed method will be compared with the existing methods in terms of Peak Signal to noise ratio (PSNR) and Mean Square Error (MSE).

Keywords: Feature Extraction, Facial Beautification, Edge Preserving Energy Minimization, Adaptive Edge Preserving Energy Minimization, Iterative steering Kernel Regression.