I am looking to build an app where users to create custom photos for themselves.
In our service the user will upload an image of themselves, use impainting to crop out parts of the image.
Then user will be able to generate very realistic looking images of themselves, (in the inpainted areas using text) for example themselves on the beach playing volleyball, or skydiving, etc. (Depends on what prompt they enter and want it to be).
All of this will be done via dreambooth using stable diffusion API.
Here is a link showing better what I want:
Also this following link performs the task I need done within Bubble PERFECTLY, only the caveat that this uses OpenAI API instead of StableDiffusionAPI.
Please keep in mind that inpainting is key for this plugin, otherwise I could easily just make API calls in Bubble to do this myself. But the inpainting task is what makes this impossible in Bubble at the moment so please hotw can I go about that?