latent_diffusion_upscaler.ipynb

Kernel: Python 3

Image super-resolution using Latent Diffusion

This colab notebook shows how to use the Latent Diffusion image super-resolution model using 🧨 diffusers libray.

The model was originally released in Latent Diffusion repo. It's a simple, 4x super-resolution model diffusion model. This model is not conditioned on text.

Install the Deps

!pip install -qq diffusers==0.11.1 accelerate

  Installing build dependencies ... done
  Getting requirements to build wheel ... done
    Preparing wheel metadata ... done
     |████████████████████████████████| 175 kB 5.1 MB/s 
     |████████████████████████████████| 182 kB 46.5 MB/s 
  Building wheel for diffusers (PEP 517) ... done

Imports

import torch
from PIL import Image
import requests
from io import BytesIO

from diffusers import LDMSuperResolutionPipeline

Load the pipeline

device = "cuda"
pipe = LDMSuperResolutionPipeline.from_pretrained( "CompVis/ldm-super-resolution-4x-openimages")
pipe = pipe.to(device)

Fetching 6 files:   0%|          | 0/6 [00:00<?, ?it/s]

Get the image for demo

# let's download an  image
url = "https://i.pinimg.com/236x/af/84/56/af8456faa55d76bd9afa18cd2fd72d58.jpg"
response = requests.get(url)
low_res_img = Image.open(BytesIO(response.content)).convert("RGB")
low_res_img = low_res_img.resize((128, 128))
low_res_img

Run pipeline to upscale the image

# run pipeline in inference (sample random noise and denoise)
upscaled_image = pipe(low_res_img, num_inference_steps=100, eta=1).images[0]

  0%|          | 0/100 [00:00<?, ?it/s]

upscaled_image