This new AI editing tool may beat Dall-E, Midjourney and Photoshop
A new AI editing tool, called DragGAN is said to replace the underlying technicalities of Photoshop with its simple drag controls.
Highlights
- DragGAN allows simple point and drag image editing
- This AI tool is speculated to perform better than Dall-E and Midjourey
- Interface of DragGAN is user-friendly and interactive
Recently, a group of Google researchers, along with Max Planck Institute of Informatics and MIT CSAIL announced DragGAN, an AI-enabled photo editing tool which enables users to edit photos way more conveniently than Photoshop.
What's different in DragGAN?
As per the recent announcement from researchers, DragGAN is said to enable users to drag and drop a point on an image, to further change its structure and entire pixels. Moreover, it’s claimed to be better than already existing popular AI image tools, Dall-E and Midjourney, which are only capable of processing highly specific prompts but not the desired poses or layouts, which is claimed to be catered by DragGAN.
his interactive technique for simple point-based picture modification uses a pre-trained GAN (generative adversarial network) to synthesise images that adhere to a variety of original images while also closely following user input. The DragGAN research article highlights the tool's primary advantage, which is its interface's simplicity and intuitiveness. Users do not need to understand the underlying technology in order to understand the functionality in a matter of seconds, stated experts.
Key features of DragGAN
The interface focuses on giving an image a starting point and an ending point. For example, in order to add a smile on a person’s face, the users just need to add two points at the corners and further two points at distance from the former ones. After dragging, the user is supposed to press the start button, which would eventually animate the image from its starting and ending points.
“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,” notes the research paper. Additionally, DragGAN provides users with a masking tool that enables them to highlight only the portions of a picture they want to change while leaving the rest alone.
Overall, this image editing tool can assist in solving the problem of the randomised nature of image generating technologies. Users will be able to produce outputs that are more similar to the image they have in mind if DragGAN is used in conjunction with image generating tools.
As of now, this tool is only a demo right now, and will soon be available to the general public.
