Diffusion Model

Detailed Explanation

Diffusion models are a class of generative models that work by gradually adding noise to data and then learning to reverse this process. The training process involves two phases: a forward diffusion process that progressively adds noise to data until it becomes pure noise, and a reverse diffusion process that learns to recover the original data by gradually removing noise. During generation, these models start with random noise and iteratively denoise it to produce high-quality samples. Diffusion models have achieved state-of-the-art results in image, audio, and video generation, becoming the foundation for many popular text-to-image systems.

Examples

• Stable Diffusion
• DALL-E
• Imagen