Particle-Filtering-based Latent Diffusion for Inverse Problems
This work addresses the problem of insufficient solution space exploration in diffusion-based inverse problem solvers for image processing, offering an incremental improvement over existing methods.
The paper tackles the limited exploration of solution spaces in latent diffusion models for inverse problems by introducing a particle-filtering-based framework, resulting in state-of-the-art performance on FFHQ-1K and ImageNet-1K datasets for tasks like super-resolution, Gaussian deblurring, and inpainting.
Current strategies for solving image-based inverse problems apply latent diffusion models to perform posterior sampling.However, almost all approaches make no explicit attempt to explore the solution space, instead drawing only a single sample from a Gaussian distribution from which to generate their solution. In this paper, we introduce a particle-filtering-based framework for a nonlinear exploration of the solution space in the initial stages of reverse SDE methods. Our proposed particle-filtering-based latent diffusion (PFLD) method and proposed problem formulation and framework can be applied to any diffusion-based solution for linear or nonlinear inverse problems. Our experimental results show that PFLD outperforms the SoTA solver PSLD on the FFHQ-1K and ImageNet-1K datasets on inverse problem tasks of super resolution, Gaussian debluring and inpainting.