A new study from MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) demonstrates a significant advancement in AI-powered image generation. The research introduces a method that allows users to make precise, localized edits to images using natural language instructions, such as ‘make the hat red’ or ‘add a tree to the left.’ This technique, which refines existing diffusion models, offers more control than previous all-or-nothing regeneration approaches. It works by identifying and modifying specific regions of an image corresponding to the text prompt while preserving the rest of the scene. The development points toward more intuitive and controllable AI tools for digital art, design, and content creation. Read the full article at: https://technologyreview.com/2024/03/14/ai-image-editing-localized-control