BriefGPT.xyz
Jun, 2024
贝叶斯条件扩散模型用于逆问题
Bayesian Conditioned Diffusion Models for Inverse Problems
HTML
PDF
Alper Güngör, Bahri Batuhan Bilecen, Tolga Çukur
TL;DR
扩散模型在图像重建中表现出优异的性能,提出了一种基于贝叶斯条件技术的扩散模型,通过条件得分函数来解决图像重建中出现的挑战性逆问题,并在图像去混叠、去模糊、超分辨率和修复中展现出最新技术的性能。
Abstract
diffusion models
have recently been shown to excel in many
image reconstruction
tasks that involve inverse problems based on a forward measurement operator. A common framework uses task-agnostic unconditional mod
→