ddpm-celebahq-finetuned-vintage-face-2epochs

This model was fine-tuned for 2 epochs on the CelebAHQ dataset to generate vintage-style face images.

Denoising Evolution (Step 0 to 50)

The image below shows the generation process from pure noise (left) to the final face (right), captured every 10 steps.

Usage

from diffusers import DDPMPipeline

pipeline = DDPMPipeline.from_pretrained('naifenn/ddpm-celebahq-finetuned-vintage-face-2epochs')
image = pipeline(num_inference_steps=50).images[0]
image