In the previous article, we have examined in depth how artificial intelligence is successful in art under the heading of Artificial Intelligence Art. In this article, we will introduce GauGAN, an artificial intelligence application that allows making beautiful paintings to anyone who does not have the ability.
Developed by Nvidia, GauGAN artificial intelligence, called Generative Adversarial Network, can turn your sketches, or rather doodles, into gorgeous pictures like pictures with a paint tool.
GauGAN Artificial Intelligence
Generative Adversarial Network (GAN)
GANs are basically composed of two neural networks that work for different purposes.
The first one (generator) tries to produce fake data like a counterfeiter, and the other (discriminator) tries to compare whether the counterfeit data presented to itself by the counterfeiter is a fake.
The discriminator, trained on real images, provides pixel-by-pixel feedback to the generator about how to improve the realism of synthetic images, creating the one closest to the real image.
Some of the outputs they produce can even be a work of art and find buyers for thousands of dollars. As an example, a picture made by artificial intelligence found a buyer for exactly 432 thousand 500 dollars.
Here you can find out more about this portrait by Edmond de Belamy, created by GAN.
Draw it, AI turns it into a masterpiece! (GauGAN)
GauGAN, created with similar neural networks, uploaded approximately 1 million photos to the system so that artificial intelligence could learn. In this way, the objects in these photos and their relationship with each other were analyzed by the software.
GauGAN AI could offer a powerful tool for creating virtual worlds to everyone from architects and urban planners to landscape designers and game developers. With an Artificial Intelligence (AI) that understands how the real world looks like, these professionals could better prototype ideas and make rapid changes to a synthetic scene.
With the GauGAN application, users are allowed to use buttons such as sand, sky, sea, and snow to scribble and change them any way they want.
For example, you draw a mountain and you think it is not beautiful. Just like using the paint tool, you can immediately put another object in its place in seconds and change your picture. Wow, it’s beautiful now!
Applying a deep learning model by analyzing 1 million images, GauGAN, easily makes a scribbled picture a beautiful masterpiece for you.
GauGAN application, which was released as a demo at the beginning of 2019 and is not available to everyone, currently focuses on natural elements such as land, sea, and sky, while it is planned to start to fill other landscape features, including buildings, roads, and people.
Can’t you draw as you imagined?
The pictures that come to your mind are flawless, but aren’t your drawing skills enough to reflect your dreams? Here’s the benefit of technology, the opportunity to realize your dream on simple lines.
Here you can watch the video of the GauGAN project.