Is it possible to use the region nodes to mimic a standard img2img workflow but centered on a specific area instead of the entire image?
For example, in this image, I want SD to make the hair longer, so I painted in some more hair. Is there a way I can make SD img2img only that area by using regions?