The experimental AI tool makes editing (editing) images a breeze with a simple click and drag gesture. The productive artificial intelligence tool, called DragGAN, allows you to make radical changes to images in just seconds. You can think of DragGAN as Photoshop’s Warp tool, but much more powerful.The productive AI race
By now even the most casual tech news follower is aware of productive AI tools like ChatGPT, Stable Diffusion, Midjourney, and DALL-E. The world’s top 5 companies compete to develop the best major language models and incorporate them into every software or web service we use. These tools can generate useful images or text using prompts. On the other hand, many of these tools are “trained” on works written by humans and require human oversight to bring their output to a meaningful level.
Revolutionizing visual editing with DragGAN
However, new artificial intelligence research reveals incredible progress, especially in the field of image manipulation. A group of scientists from Google, MIT, the University of Pennsylvania, and the Max Planck Informatics Institute in Germany have developed an experimental tool that could make image editing easier and more accessible to ordinary people.
It is enough to look at the examples in this news to understand what the new tool called DragGAN can do. With just a few clicks and a few seconds, it is possible to rotate the object in the image as if it were a 3D model, change facial expressions or make any other difficult adjustments you can think of. In the meantime, it should be noted that DragGAN is not a public model. Therefore, we did not have the opportunity to try the tool.
Your dream scene is just seconds away
The researchers note that DragGAN can change the content of an image in just a few seconds when using Nvidia’s GeForce RTX 3090 graphics card, because their application does not need to use multiple neural networks to achieve the desired results. The next step will be to develop a similar model for point-based editing of 3D models.
However, as stated by the relevant research team, the most interesting aspect of this study is not the image manipulation, but the user interface. We’ve been able to use artificial intelligence tools like GANs to create realistic images for some time, but most methods lack flexibility and precision. Ask an AI image generator to “make a picture of a lion prowling through the savanna,” and you’ll get one, but it may not be exactly the pose you want or need. Even AI prompt engineering has sprung up to get the maximum out of these prompts and can earn more than $300,000 a year. DragGAN makes it possible to create your dream scene without the need for these detailed prompts.
27254
