𝗗𝗶𝗳𝗳𝘂𝘀𝗶𝗼𝗻 𝗠𝗼𝗱𝗲𝗹𝘀 𝗳𝗼𝗿 𝗜𝗻𝘃𝗲𝗿𝘀𝗲 𝗣𝗿𝗼𝗯𝗹𝗲𝗺𝘀

𝗗𝗶𝗳𝗳𝘂𝘀𝗶𝗼𝗻 𝗠𝗼𝗱𝗲𝗹𝘀 𝗳𝗼𝗿 𝗜𝗻𝘃𝗲𝗿𝘀𝗲 𝗣𝗿𝗼𝗯𝗹𝗲𝗺𝘀

Inverse problems happen when you have damaged data. You want to find the original image.

Examples include:

• Removing noise.
• Fixing blur.
• Filling missing pixels.

Diffusion models help here. They learn to remove noise step by step. They turn random noise into clear pictures. This method gives high quality results. It works across different data types.

Read the full survey to learn more.

Source: https://dev.to/paperium/a-survey-on-diffusion-models-for-inverse-problems-pbo Optional learning community: https://t.me/GyaanSetuAi