An Unsupervised Approach towards Varying Human Skin Tone Using Generative Adversarial Networks

With the increasing popularity of augmented and virtual reality, retailers\nare now focusing more towards customer satisfaction to increase the amount of\nsales. Although augmented reality is not a new concept but it has gained much\nneeded attention over the past few years. Our present work is targeted towards\nthis direction which may be used to enhance user experience in various virtual\nand augmented reality based applications. We propose a model to change skin\ntone of a person. Given any input image of a person or a group of persons with\nsome value indicating the desired change of skin color towards fairness or\ndarkness, this method can change the skin tone of the persons in the image.\nThis is an unsupervised method and also unconstrained in terms of pose,\nillumination, number of persons in the image etc. The goal of this work is to\nreduce the time and effort which is generally required for changing the skin\ntone using existing applications (e.g., Photoshop) by professionals or novice.\nTo establish the efficacy of this method we have compared our result with that\nof some popular photo editor and also with the result of some existing\nbenchmark method related to human attribute manipulation. Rigorous experiments\non different datasets show the effectiveness of this method in terms of\nsynthesizing perceptually convincing outputs.\n