A Gradient Flow Approach to Solving Inverse Problems with Latent Diffusion Models
This addresses the problem of solving inverse problems for researchers and practitioners in fields like imaging or data analysis, but it appears incremental as it builds on existing diffusion models.
The paper tackled solving ill-posed inverse problems by proposing a training-free method called Diffusion-regularized Wasserstein Gradient Flow (DWGF) that leverages pretrained latent diffusion models, demonstrating performance on standard benchmarks using StableDiffusion as the prior.
Solving ill-posed inverse problems requires powerful and flexible priors. We propose leveraging pretrained latent diffusion models for this task through a new training-free approach, termed Diffusion-regularized Wasserstein Gradient Flow (DWGF). Specifically, we formulate the posterior sampling problem as a regularized Wasserstein gradient flow of the Kullback-Leibler divergence in the latent space. We demonstrate the performance of our method on standard benchmarks using StableDiffusion (Rombach et al., 2022) as the prior.