MGIE can handle a wide range of editing scenarios, from simple color adjustments to complex object manipulations. It can perform common Photoshop-style edits, optimize the overall quality of a photo, and edit specific regions or objects in an image. The model is available as an open-source project on GitHub, and can also be tried out online through a web demo hosted on Hugging Face Spaces. The release of MGIE is seen as a breakthrough in the field of instruction-based image editing and a demonstration of Apple's growing prowess in AI research and development.
Key takeaways:
- Apple has released a new open-source AI model, MGIE (MLLM-Guided Image Editing), which can edit images based on natural language instructions.
- MGIE uses multimodal large language models (MLLMs) to interpret user commands and perform pixel-level manipulations, handling various editing aspects such as Photoshop-style modification, global photo optimization, and local editing.
- The model is the result of a collaboration between Apple and researchers from the University of California, Santa Barbara, and was presented at the International Conference on Learning Representations (ICLR) 2024.
- MGIE is available as an open-source project on GitHub, and can be integrated with other applications or platforms that require image editing functionality.