Neural Design Space Exploration

We have developed a new tool that makes it possible for people with zero programming experience to intentionally and meaningfully explore the latent space of a GAN. This tool was developed to provide a means for designers to explore the "design space" of their domains. While the latent space of a GAN may be capable of capturing a domain’s design space, there is still the problem of exploring that space. These spaces contain hundreds or thousands of dimensions which are typically entangled. Furthermore, while generating random samples is trivial, intentionally locating a starting point within the latent space based on a target image is not. Our tool provides technical solutions to these problems, and packages them in a web-based graphical interface that is designed for those with no programming experience.