Bumblebee and Stable Diffusion Image to Image Generation

I am looking to use Hugging Face’s Stable Diffusion Image to Image model to create some images, but I am not sure how to go about this using Bumblebee. I see that there are functions for text_to_image and image_classification, but I haven’t figured out how to combine the two (?) to create new images based on an input image and a prompt. Has anyone done this before? If you have, I could use some help. Thanks in advance.

2 Likes

see:

3 Likes

Very helpful, thanks!!

1 Like

This is awesome! It is making me want to do that project I have been meaning to do for a while!

I hope Alekx (@ElixirCasts) or someone else will create a screencast on getting up and running with this along with a few demonstrations of what’s possible with it and maybe @kip can find some way to integrate it with his Image library :sunglasses:

I have added #bumblebooth to the tags for this thread as I reckon a lot of people will be interested in it!

3 Likes

it’s by @hanrelan there is also a branch with further progress towards “dreambooth” - but I’m by all means a noob in this domain…

2 Likes

Thanks Peter - I am too! :lol:

Please keep us posted @hanrelan :blush:

2 Likes