From posterior sampling to meaningful diversity in image restoration
Image restoration problems are typically ill-posed in the sense that each degraded image
can be restored in infinitely many valid ways. To accommodate this, many works generate a …
can be restored in infinitely many valid ways. To accommodate this, many works generate a …
Diffusion-based image inpainting with internal learning
N Cherel, A Almansa, Y Gousseau… - arXiv preprint arXiv …, 2024 - arxiv.org
Diffusion models are now the undisputed state-of-the-art for image generation and image
restoration. However, they require large amounts of computational power for training and …
restoration. However, they require large amounts of computational power for training and …