Image Inpainting via Tractable Steering of Diffusion Models

TL;DR

This paper introduces a novel method for image inpainting that uses tractable probabilistic models to precisely steer diffusion models, improving image quality and semantic coherence with minimal additional computation.

Contribution

It leverages Probabilistic Circuits to guide diffusion models for constrained inpainting, enabling exact posterior computation and enhanced control over generated images.

Findings

Improved inpainting quality across multiple datasets.

Achieved semantic coherence with only ~10% extra computation.

Enabled region-specific semantic constraints in image generation.

Abstract

Diffusion models are the current state of the art for generating photorealistic images. Controlling the sampling process for constrained image generation tasks such as inpainting, however, remains challenging since exact conditioning on such constraints is intractable. While existing methods use various techniques to approximate the constrained posterior, this paper proposes to exploit the ability of Tractable Probabilistic Models (TPMs) to exactly and efficiently compute the constrained posterior, and to leverage this signal to steer the denoising process of diffusion models. Specifically, this paper adopts a class of expressive TPMs termed Probabilistic Circuits (PCs). Building upon prior advances, we further scale up PCs and make them capable of guiding the image generation process of diffusion models. Empirical results suggest that our approach can consistently improve the overall quality and semantic coherence of inpainted images across three natural image datasets (i.e., CelebA-HQ, ImageNet, and LSUN) with only $\sim\! 10 \%$ additional computational overhead brought by the TPM. Further, with the help of an image encoder and decoder, our method can readily accept semantic constraints on specific regions of the image, which opens up the potential for more controlled image generation tasks. In addition to proposing a new framework for constrained image generation, this paper highlights the benefit of more tractable models and motivates the development of expressive TPMs.