Interactive GAN (iGan) is an interactive image generation interface. It is a system based on deep generative models, Generative Adversarial Networks (GAN) and CDGAN.
iGAn is an intelligent drawing interface, it automatically generates images inspired by the color and shape of the brush strokes. It is an interactive visual debugging tool for understanding and visualizing deep generative models. It interacts with the generative model and can understand what visual content the model can produce, as well as the limitation of the model.
Generative Visual Manipulationon the Natural Image Manifold
Jun-Yan Zhu, Philipp Krahenbuhl, Eli Shechtman and Alexei Efros