Apple has just introduced a groundbreaking tool named MGIE, a collaboration with the University of California, Santa Barbara. This new model allows users to describe photo edits using plain language, eliminating the need for hands-on editing software.
MGIE, short for MLLM-Guided Image Editing, can perform tasks such as cropping, resizing, flipping, and applying filters based on text prompts.
Unlike other models, MGIE can handle both simple and complex editing requests. It interprets user prompts and then “imagines” the desired edit. For instance, asking for a bluer sky results in the model enhancing the brightness of the sky in the image. Apple has made MGIE available for download on GitHub and also provided a web demo on Hugging Face Spaces.
While Apple has not revealed its future plans for MGIE, this move aligns with CEO Tim Cook’s statement about incorporating more AI features into Apple devices. The company has been relatively less involved in the generative AI space compared to Microsoft, Meta, or Google.
ALSO:
In a world where image generation platforms and AI editing models are gaining prominence, Apple’s MGIE emerges as a user-friendly and efficient solution for text-driven photo edits.
