אירועים והרצאות בפקולטה למדעי המחשב ע"ש הנרי ומרילין טאוב

Plug-and-play methods for solving inverse problems have continuously improved over the years by incorporating more advanced image priors. Latent diffusion models are among the most powerful priors, making them a natural choice for solving inverse problems.  

However, existing approaches require multiple applications of an Autoencoder to transition between pixel and latent spaces during restoration, leading to high computational costs and degraded restoration quality.  

In this work, we introduce a new plug-and-play paradigm that operates entirely in the latent space of diffusion models. By emulating pixel-space degradations directly in the latent space through a short learning phase, we eliminate the need for the Autoencoder during restoration, enabling faster inference and improved restoration fidelity.

We validate our method across various image restoration tasks and datasets, achieving significantly higher perceptual quality than previous methods while being 2.6-10 times faster in inference and 1.7-7 times faster when accounting for the learning phase of the latent operator.