Stock photos are the bane of many marketers and designers. Image-heavy providers are prohibitively expensive, while the more affordable ones might result in a worried call from HR.
At TNW Towers, we gently bypass these barriers by creating custom masterpieces, like these beautiful photos of Mark Zuckerberg:
Unfortunately, not everyone is endowed with such virtuosity, including yours. Take a look at the monstrosity I made at the top of this article.
Fortunately, AI may soon cover up our artistic incompetence, by enabling the generation of photorealistic images.
The new from Nvidia GauGAN2 demo gives a glimpse of the potential. the system can translate text into realistic images, which you can then edit to your heart’s content.
The model, which is powered by generative adversarial networks (GANs), combines segmentation mapping, inpainting and text-to-image generation.
You can use these features to create something magical, like unicorns and rainbows, or something more realistic, like the apocalypse.
The demo works best with simple descriptions of nature, such as “sunset on a rocky beach” or “snow in the mountains.” More complicated text can produce weird output, but it gives the system a chance to show off its editing power.
You can generate a segmentation map of objects in the scene, then switch to drawing mode to doodle the adjustments. GauGAN2’s smart brush will then convert the sketches into the style of a photo.
This feature lets you make realistic edits, like adding a tree or making a mountain bigger. It can also be used to create otherworldly imagery.
“Imagine, for example, recreating a landscape of the iconic planet of Tatooine in the star wars franchise, which has two suns,” Nvidia’s Ishan Salian said in a blog post.
“All that’s needed is the text ‘desert hills sun’ to create a starting point, after which users can quickly draw a second sun.”
The results aren’t always perfect, but the approach has enormous potential. It might not be long before we ditch stock photos for an infinitely customizable AI image generator.