There's a new photo editor from Apple, if you know where to look for it. iPhone Kings has teamed up with researchers at the University of California, Santa Barbara to build a tool that lets you edit photos and images using text instructions. It doesn't have an official release, but researchers are hosting a demo you can try for yourself, which has been spotted for the first time Extreme technology.
The project is called Multimodal Large Language Model Guided Image Editing (MGIE). There are a lot of AI photo editing software on the market today. Photoshop now comes with built-in AI tools, and others like OpenAI's DALL-E allow you to edit images as well as create them from whole cloth. However, if you've tried to use them before, you know that it can be a bit frustrating. In many cases, AI has difficulty understanding exactly what you are looking for.
The innovation at MGIE adds another layer of AI interpretation. When you tell the AI what you want to see, MGIE first uses text-based AI to make your instructions clearer and more descriptive. “Experimental results show that expressive instructions are essential for instruction-based image editing,” the researchers said in an article. paper Published on arXiv. “MGIE can lead to significant improvement.”
Apple has published an open source version of the software on github. If you're smart, you can get a copy of MGIE running yourself, but the researchers set up the tool on… Face hugging. It runs a little slow when there are a lot of people using it, but it's a fun experience.
Tech giants like Apple are spending billions of dollars on projects that no one will ever see, so it's entirely possible that the so-called MGIE tool will never get an official release. Apple did not immediately respond to a request for comment.
We took it for a spin here at the Gizmodo office. I uploaded a photo of my colleague and closest advisor Kyle Barr wearing weird sunglasses that I took on Netflix in… Consumer Electronics Show this year. I told Amnesty International: “The man is standing in the desert.” Before creating the image, MGIE extrapolates the following:
“The man is wearing a metal helmet and standing in a desert environment. The surrounding environment is barren and barren, with sand dunes stretching as far as the eye can see.”
After playing around with the tool for a lot longer than it should have, it's clear that it's subject to many of the same limitations as any other AI image generator. Often times, the results are strange and nothing like what you ordered. But in some cases, it did an impressive job, and in the software's defense, the AI performed better with familiar topics. “Familiar” is not something you would call Kyle's sunglasses.
