Anybody can now learn Photoshop without having to worry about the underlying technicalities, thanks to a new tool that harnesses the power of generative AI. DragGAN is a user-friendly application that lets anyone significantly alter photos using straightforward point-and-drag controls.
DragGAN allows users to drop a point on a picture and change the structure and entire pixels, as described in a paper by researchers from Google, the Max Planck Institute of Informatics, and the MIT CSAIL. This distinguishes it from other well-known generative AI image programmes like Dall-E and Midjourney, which, despite being able to comprehend extremely particular cues, are unable to exactly create specified poses or layouts.
Examples in the study include a photo of an automobile that has been altered to look like it was taken from an entirely other angle and a mountain that has been stretched to double its original height. The strength of generative AI allows the image to look realistic despite such substantial modifications.
Read More: OpenAI Closes $300 Million Funding Round Between $27-$29 billion Valuation
The DragGAN research article highlights the tool’s primary advantage, which is its interface’s simplicity and intuitiveness, in addition to its amazing capabilities. Users do not need to understand the underlying technology in order to understand the functionality in a matter of seconds.
The interface focuses on giving an image a starting point and an ending point. Users can place two points at the corners of the mouth and two more points that are a little further away, for instance, to make someone smile. When you press the Start button, the mouth of the tool stretches animatedly from the start points to the end points.
In the meanwhile, generative AI fills in any gaps that may occur while maintaining realism. “Our approach can hallucinate occluded content, like the teeth inside a lion’s mouth, and can deform following the object’s rigidity, like the bending of a horse leg,” according to the research article. Additionally, DragGAN has a masking tool that enables users to draw attention to only the portions of a picture they want to change while leaving the rest alone.
