I need someone who can help me with Stable Diffusion.
I'm trying to use it to create realistic variations of a headshot photo. I've been experimenting with ControlNet and various models and I've read that I should be able to get realistic looking image variations using this method.
But, it's not working for me: The image outputs are significantly different than to image input. i.e. the person looks like a different person.
My desired outcome:
Upload a photo of a person (me for example), the use Stable Diffusion to generate new photos of me in different scenarios e.g studio headshot, headshot with blurred city background etc. The new images look like the original person.
I have installed Stanbel Diffusionad various models + controlNet on my Mac.
And I have been using a hosted version of Stable Diffusion using RunDiffusion.
Please ler em know if you can help with this and can demo how to do it.
