An Expectation-Maximization Algorithm for Training Clean Diffusion Models from Corrupted Observations
Diffusion models excel in solving imaging inverse problems due to their ability to model complex image priors. However, their reliance on large, clean datasets for training limits their practical use where clean data is scarce. In this paper, we propose EMDiffusion, an expectation-maximization (EM) approach to train diffusion models from corrupted observations. Our method alternates between reconstructing clean images from corrupted data using a known diffusion model (E-step) and refining diffusion model weights based on these reconstructions (M-step). This iterative process leads the learned diffusion model to gradually converge to the true clean data distribution. We validate our method through extensive experiments on diverse computational imaging tasks, including random inpainting, denoising, and deblurring, achieving new state-of-the-art performance.
Code (0)
등록된 구현이 없습니다.
Tasks
DeblurringDenoisingMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
Learning Diffusion Priors from Observations by Expectation Maximization
Diffusion models recently proved to be remarkable priors for Bayesian inverse problems. However, training these models typically requires access to large amounts of clean data, which could prove difficult in some setting…
Learning Diffusion Model from Noisy Measurement using Principled Expectation-Maximization Method
Diffusion models have demonstrated exceptional ability in modeling complex image distributions, making them versatile plug-and-play priors for solving imaging inverse problems. However, their reliance on large-scale clea…
DeblurringDenoisingDiffEM: Learning from Corrupted Data with Diffusion Models via Expectation Maximization
Diffusion models have emerged as powerful generative priors for high-dimensional inverse problems, yet learning them when only corrupted or noisy observations are available remains challenging. In this work, we propose a…
Image ReconstructionScore-based diffusion priors for multi-target detection
Multi-target detection (MTD) is the problem of estimating an image from a large, noisy measurement that contains randomly translated and rotated copies of the image. Motivated by the single-particle cryo-electron microsc…
Learning with Noisy Labels: Interconnection of Two Expectation-Maximizations
Labor-intensive labeling becomes a bottleneck in developing computer vision algorithms based on deep learning. For this reason, dealing with imperfect labels has increasingly gained attention and has become an active fie…
Learning with noisy labels