Sometimes I have a picture but my own description doesn't produce a result similar to it.
I want to be able to generate a prompt based on an image so that the generation is closer to what I need.