Cartoon Character Generation using Generative Adversarial Network

Abstract

Animated faces show up in cartoons, comics and games. They are broadly utilized as profile pictures in online life stages, for example, Facebook and Instagram. Drawing an animated face is work intense. Not just it requires proficient skills but also its time consuming. A lot of time is wasted in creating a cartoon character from scratch, and most of the cases ends up in creating an awkward character having very low polygon intensity. Generative adversarial network (GAN) framework can be trained with a collection of cartoon images. GANs comprises of a generator network and a discriminator network. Because of the ability of deep networks and the competitive training algorithm, GANs produce realistic images, and have great potential in the field of image processing. This method turns out to be a surprisingly handy tool in enhancing blurry images. The underlying idea behind GAN is that it contains two neural networks that compete with each other in a zero-sum game, which constitutes of generator and a discriminator.