Apple made an AI image tool that lets you make edits by describing them

Date:

Share:


Apple researchers released a new model that lets users describe in plain language what they want to change in a photo without ever touching photo editing software.

The MGIE model, which Apple worked on with the University of California, Santa Barbara, can crop, resize, flip, and add filters to images all through text prompts. 

MGIE, which stands for MLLM-Guided Image Editing, can be applied to simple and more complex image editing tasks like modifying specific objects in a photo to make them a different shape or come off brighter. The model blends two different uses of multimodal language models. First, it learns how to interpret user prompts. Then it “imagines” what the edit would look like (asking for a bluer sky in a photo becomes bumping up the brightness on the sky portion of an image, for example).

When editing a photo with MGIE, users just have to type out what they want to change about the picture. The paper used the example of editing an image of a pepperoni pizza. Typing the prompt “make it more healthy” adds vegetable toppings. A photo of tigers in the Sahara looks dark, but after telling the model to “add more contrast to simulate more light,” the picture appears brighter. 

Screenshot of the MGIE paper.
Image: Apple

“Instead of brief but ambiguous guidance, MGIE derives explicit visual-aware intention and leads to reasonable image editing. We conduct extensive studies from various editing aspects and demonstrate that our MGIE effectively improves performance while maintaining competitive efficiency. We also believe the MLLM-guided framework can contribute to future vision-and-language research,” the researchers said in the paper. 

Apple made MGIE available through GitHub for download, but it also released a web demo on Hugging Face Spaces, reports VentureBeat. The company did not say what its plans for the model are beyond research.

Some image generation platforms, like OpenAI’s DALL-E 3, can perform simple photo editing tasks on pictures they create through text inputs. Photoshop creator Adobe, which most people turn to for image editing, also has its own AI editing model. Its Firefly AI model powers generative fill, which adds generated backgrounds to photos. 



Source link

Subscribe to our magazine

━ more like this

Young British men are NEETs—not in employment, education, or training—more than women

The number of young British men who are neither in work nor preparing for the world of work is at its highest in...

Flappy Bird’s original creator says he has nothing to do with the new game

Last week, The Flappy Bird Foundation announced a game called Flappy Bird. But while the group has been framing it as the triumphant...

VCs backing Kamala Harris are ‘trying to buy the right type of risk’

Being a venture capitalist carries a lot of prestige in Silicon Valley. Those who choose which startups to fund see themselves as fostering...

SpaceX capsule returns to Earth after historic spacewalk that went higher than ISS, Hubble

A billionaire spacewalker returned to Earth with his crew on Sunday, ending a five-day trip that lifted them higher than anyone has traveled since NASA’s...

The great Evernote reboot

For so many years, the Evernote elephant was a truly iconic logo. Evernote was one of the first productivity apps to embrace smartphones,...