Sketch To Photo Synthesis

Abstract

Face photo-sketch synthesis aims at generating a face sketch (or photo) from an input face photo (or sketch). It is of wide applications ranging from law enforcement to digital entertainment. For instance, the synthesized facial photos can be utilized to identify the suspects when only the sketches of mug-shots are available. Face sketch synthesis phenomenon, a kind of image-image translation, generates synthesized face/sketch with wide range of applications pertaining law enforcement and entertainment to mention few. Despite the compelling results produced by many existing methods of late, there are still challenges due to deformation and blurred effects on facial components resulting in unrealistic face/sketch. Face sketch synthesis has made significant progress with the development of deep neural networks in these years. The delicate depiction of sketch portraits facilitates a wide range of applications like digital entertainment and law enforcement. Moreover, Generative Adversarial Network (GAN) based deep learning approaches have paved way for dealing with issues of training quality. However, accurate and realistic face sketch generation is still a challenging task due to the illumination variations and complex backgrounds in the real scenes. This research is aimed at proposing a novel framework based on deep Generative Adversarial Networks to enhance face sketch synthesis by overcoming problems such as deformations and blurring effects in the generated artifacts.