BayesDiff: Estimating Pixel-wise Uncertainty in Diffusion via Bayesian Inference
This addresses the challenge of sample-wise quality assessment in diffusion models for image generation, though it appears incremental as it builds on existing Bayesian methods.
The paper tackled the problem of identifying low-quality generations in diffusion models by proposing BayesDiff, a pixel-wise uncertainty estimator based on Bayesian inference, which achieved effective filtering of low-fidelity images and improved text-to-image tasks.
Diffusion models have impressive image generation capability, but low-quality generations still exist, and their identification remains challenging due to the lack of a proper sample-wise metric. To address this, we propose BayesDiff, a pixel-wise uncertainty estimator for generations from diffusion models based on Bayesian inference. In particular, we derive a novel uncertainty iteration principle to characterize the uncertainty dynamics in diffusion, and leverage the last-layer Laplace approximation for efficient Bayesian inference. The estimated pixel-wise uncertainty can not only be aggregated into a sample-wise metric to filter out low-fidelity images but also aids in augmenting successful generations and rectifying artifacts in failed generations in text-to-image tasks. Extensive experiments demonstrate the efficacy of BayesDiff and its promise for practical applications.