CVNov 18, 2022

A Structure-Guided Diffusion Model for Large-Hole Image Completion

arXiv:2211.10437v34 citationsh-index: 101Has Code
Originality Highly original
AI Analysis

This addresses the problem of generating plausible image completions for large missing regions, which is incremental as it builds on existing diffusion models with structural guidance.

The paper tackles the challenge of large-hole image completion by integrating explicit structural guidance into a diffusion model, achieving superior or comparable visual quality on Places and CelebA-HQ datasets compared to state-of-the-art methods.

Image completion techniques have made significant progress in filling missing regions (i.e., holes) in images. However, large-hole completion remains challenging due to limited structural information. In this paper, we address this problem by integrating explicit structural guidance into diffusion-based image completion, forming our structure-guided diffusion model (SGDM). It consists of two cascaded diffusion probabilistic models: structure and texture generators. The structure generator generates an edge image representing plausible structures within the holes, which is then used for guiding the texture generation process. To train both generators jointly, we devise a novel strategy that leverages optimal Bayesian denoising, which denoises the output of the structure generator in a single step and thus allows backpropagation. Our diffusion-based approach enables a diversity of plausible completions, while the editable edges allow for editing parts of an image. Our experiments on natural scene (Places) and face (CelebA-HQ) datasets demonstrate that our method achieves a superior or comparable visual quality compared to state-of-the-art approaches. The code is available for research purposes at https://github.com/UdonDa/Structure_Guided_Diffusion_Model.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes