2024-02-09 01:00:00
Written by Amira Shehata Friday, February 9, 2024 03:00 AM Issued Apple company A new open source AI model for image editing is called MLLM Guided Image Editing (MGIE), which uses multimodal large language models (MLLMs) to interpret text-based commands when processing images.
According to what was reported by Engadget, the tool has the ability to edit images based on the text that the user writes.
Although it’s not the first tool that can do this, “human instructions are sometimes too brief for existing methods to capture and follow,” the project says.
The company developed MGIE with researchers from the University of California, California, which has the ability to transform simple or ambiguous text prompts into more detailed and clear instructions that the image editor himself can follow.
In addition to making significant changes to images, MGIE can also crop, resize and rotate images, as well as improve brightness, contrast and color balance, all through text prompts.
It can also modify specific areas of the image and can, for example, modify the hair, eyes and clothing of the person in it, or remove objects in the background.
Apple released the model through GitHub, but those interested can also try out the demo, which is currently hosted on Hugging Face Spaces.
Apple has not yet made clear whether it plans to use what it learned from this project in a tool or feature that it could integrate into any of its products.
#Apple #model #edit #photos #based #text #commands