DDPM — Denoising Diffusion Probabilistic Model

Fact

A diffusion model that makes data by removing noise step by step.

In Plain Words

DDPM starts with a screen full of TV static. Then it cleans the mess, step by step, until a puppy appears.

It often powers text-to-image tools and photo edits. Results are steady, but it needs many steps.

Related Concepts

• Diffusion Model
• Generative Model
• U-Net Image Segmentation Network
• Classifier-Free Guidance

Diffusion
DDPM is a classic Diffusion model.

Generative Model
DDPM learns data patterns and makes new samples.

U-Net
U-Net often acts as the denoising network in DDPM.

CFG
CFG often guides DDPM toward the prompt.