Apple has recently unveiled ‘MGIE,’ an open-source AI model that aims to revolutionize image editing using multimodal large language models (MLLMs) (via VentureBeat).
The acronym MGIE stands for MLLM-Guided Image Editing, highlighting its reliance on MLLMs to comprehend and execute editing commands expressed in natural language.
Developed in partnership with researchers from the University of California, Santa Barbara, MGIE represents a significant advancement in instruction-based image manipulation.
This groundbreaking model was recently showcased at the prestigious International Conference on Learning Representations (ICLR) 2024, solidifying its position at the forefront of AI research.
What sets MGIE apart is its ability to interpret user instructions with exceptional precision, enabling pixel-level adjustments that cater to a wide range of editing needs. From basic Photoshop-style modifications to advanced global optimizations and intricate local edits, MGIE promises both versatility and efficiency.
One of MGIE’s key strengths is its use of MLLMs to generate clear and concise editing instructions, ensuring seamless communication between users and the AI system.
Whether it’s enhancing color saturation, refining image composition, or applying artistic effects, MGIE’s user-friendly interface streamlines the editing process for both beginners and experienced professionals.
Furthermore, MGIE’s advanced capabilities go beyond traditional image editing tasks, encompassing complex manipulations such as object removal, background replacement, and image blending. Its robust framework, supported by state-of-the-art machine learning techniques, empowers users to unleash their creativity without limitations.
For those interested in exploring the potential of MGIE, Apple has made the model available as an open-source project on GitHub. Additionally, Apple has provided a demo notebook showcasing MGIE’s capabilities along with an online web demo hosted on Hugging Face Spaces, a platform for sharing and collaborating on machine learning (ML) projects.
