Let's assume that SD v2.1 gives me realistic images and all, but I want each of my images to be centered and isolated, meaning alone and possibly with a sense of depth or perspective and moreover just the total object in the image. What should I do in this case?
Think fine tuning? Train LDM with prepared data according to that? I'm just new and i need help.
Let's assume that SD v2.1 gives me realistic images and all, but I want each of my images to be centered and isolated, meaning alone and possibly with a sense of depth or perspective and moreover just the total object in the image. What should I do in this case?
Think fine tuning? Train LDM with prepared data according to that? I'm just new and i need help.