DragGAN is a new AI app that enables you to easily adjust photos and art by simply dragging across the image. It really is that simple, and honestly, I’m a little obsessed with it.
We’ve seen how Diffusion apps like DALLE.2, Stable Diffusion, and Midjourney can scare us with AI house party images and fake beer ads, but now we’re starting to see uses for GANs (generative adversarial networks), and these are even more useful. A bit, if DragGAN is anything to go by.
While the AI photo editing software is still in its infancy – DragGAN exists as a white paper The results are impressive. DragGAN is an interactive way to edit images or artwork by marking points on the image and just dragging. Artificial intelligence does all the hard work.
The application uses artificial intelligence to process the images. There are some technical terms used but mainly the AI enables you to influence the image while maintaining its integrity; DragGAN basically allows you to distort images with interactive points, and generative AI assists with the changes for a smooth and precise end result.
You can adjust photos into different categories, including animals, cars, people, as well as landscapes, adjust everything from the layout of the photo to shapes and forms, and even change the expression of the person in the photo.
The research team wrote: “Imagine being able to ‘drag’ any point of the image to exactly where you want it. This is what we aim to achieve with our new method, which we call DragGAN.”
They explain how it works: “DragGAN has two main parts. The first is feature-based motion control that helps you move any point in the image to the target position. The second part is a new point-tracking approach that helps track the position of these points. With DragGAN, anyone can Image distortion and precise control over where each pixel ends.”
The GAN-based tool currently works with 2D images but the team plans to release a version that works with 3D models. If you’d like to learn more about upcoming AI applications, read our article on Adobe Firefly and to understand how creators of AI art are affecting artists, read our feature What Could Stop Generative AI Art?
“Typical beer trailblazer. Hipster-friendly web buff. Certified alcohol fanatic. Internetaholic. Infuriatingly humble zombie lover.”