import torch
from diffusers import AutoPipelineForInpainting
from diffusers.utils import load_image
# switch to "mps" for apple devices
pipe = AutoPipelineForInpainting.from_pretrained("carsonkatri/stable-diffusion-2-depth-diffusers", dtype=torch.float16, device_map="cuda")
img_url = "https://raw.githubusercontent.com/CompVis/latent-diffusion/main/data/inpainting_examples/overture-creations-5sI6fQgYIuo.png"
mask_url = "https://raw.githubusercontent.com/CompVis/latent-diffusion/main/data/inpainting_examples/overture-creations-5sI6fQgYIuo_mask.png"
image = load_image(img_url).resize((1024, 1024))
mask_image = load_image(mask_url).resize((1024, 1024))
prompt = "a tiger sitting on a park bench"
generator = torch.Generator(device="cuda").manual_seed(0)
image = pipe(
prompt=prompt,
image=image,
mask_image=mask_image,
guidance_scale=8.0,
num_inference_steps=20, # steps between 15 and 30 work well for us
strength=0.99, # make sure to use `strength` below 1.0
generator=generator,
).images[0]This model is converted from stable-diffusion-2-depth using the conversion script from 🤗 Diffusers.
You can use it with the pipeline here: https://gist.github.com/carson-katri/f51532b9d5162928d5cacbaee081a799
# class StableDiffusionDepthPipeline: ...
from PIL import Image
model_id = "carsonkatri/stable-diffusion-2-depth-diffusers"
# Use the pipeline from this GH Gist: https://gist.github.com/carson-katri/f51532b9d5162928d5cacbaee081a799
pipe = StableDiffusionDepthPipeline.from_pretrained(model_id)
pipe = pipe.to("cuda")
image = pipe(
prompt="a photo of a stormtrooper from star wars",
depth_image=Image.open('depth.png'), # Black and white depth map
image=Image.open("emad.png"), # Optional init image and strength.
width=768,
height=512
).images[0]
image.save('stormtrooper.png')
- Downloads last month
- 37