“Diffusion Model-Based Image Editing: A Survey” presents a comprehensive examination of diffusion models and their transformative effect on image editing. From theoretical underpinnings to practical applications, this survey traverses the landscape of image manipulation using diffusion models, encompassing learning strategies, user-input conditions, and specific editing tasks.

• An extensive survey on diffusion model-based image editing.
• Covers advancements in image inpainting, outpainting, and more.
• Highlights traditional and multimodal conditional methods.
• Introduces EditEval benchmark with LMM Score metric for performance evaluation.
• Explores future directions for the integration of text guidance in image editing.

The study captures a crucial phase in the evolution of image editing technology and offers a platform for assessing the capabilities of text-guided algorithms, potentially fostering better user experiences in various image editing software. Read more