Amortized Guidance for Image Inpainting with Pretrained Diffusion Models

• URL: http://arxiv.org/abs/2605.13010v1
• Date: Wed, 13 May 2026 05:02:47 GMT
• Title: Amortized Guidance for Image Inpainting with Pretrained Diffusion Models
• Abstract summary: We study image inpainting with generative diffusion models.<n>We introduce a middle-ground model, termed Amortized Inpainting with Diffusion.<n>AID keeps a pretrained diffusion backbone fixed, trains a small reusable guidance module offline, and then reuses it across masked images without per-instance optimization.
• Score: 3.9819516108444115
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We study image inpainting with generative diffusion models. Existing methods typically either train dedicated task-specific models, or adapt a pretrained diffusion model separately for each masked image at deployment. We introduce a middle-ground model, termed Amortized Inpainting with Diffusion (AID), which keeps a pretrained diffusion backbone fixed, trains a small reusable guidance module offline, and then reuses it across masked images without per-instance optimization. We formulate it as a deterministic guidance problem with a supervised terminal objective. To make this problem learnable in high dimensions, we derive an auxiliary Gaussian formulation and prove that solving this randomized problem recovers the optimal deterministic guidance field. This bridge yields a principled continuous-time actor--critic algorithm for learning the guidance module in a fully data-driven manner. Empirically, on AFHQv2 and FFHQ under the pixel EDM pipeline and on ImageNet under the latent EDM2 pipeline, AID consistently improves the quality--speed trade-off over strong fixed-backbone and amortized inpainting baselines across multiple mask types, while adding less than one percent trainable overhead.

Related papers

• Diffusion Models for Solving Inverse Problems via Posterior Sampling with Piecewise Guidance [52.705112811734566]
  A novel diffusion-based framework is introduced for solving inverse problems using a piecewise guidance scheme.<n>The proposed method is problem-agnostic and readily adaptable to a variety of inverse problems.<n>The framework achieves a reduction in inference time of (25%) for inpainting with both random and center masks, and (23%) and (24%) for (4times) and (8times) super-resolution tasks.
  arXiv  Detail & Related papers  (2025-07-22T19:35:14Z)
• LeDiFlow: Learned Distribution-guided Flow Matching to Accelerate Image Generation [1.1847464266302488]
  Flow Matching (FM) is a powerful generative modeling paradigm based on a simulation-free training objective instead of a score-based one used in DMs.<n>We present Learned Distribution-guided Flow Matching (LeDiFlow), a novel scalable method for training FM-based image generation models.<n>Our method utilizes a State-Of-The-Art (SOTA) transformer architecture combined with latent space sampling and can be trained on a consumer workstation.
  arXiv  Detail & Related papers  (2025-05-27T05:07:37Z)
• Steering Rectified Flow Models in the Vector Field for Controlled Image Generation [53.965218831845995]
  Diffusion models (DMs) excel in photorealism, image editing, and solving inverse problems, aided by classifier-free guidance and image inversion techniques.<n>Existing DM-based methods often require additional training, lack generalization to pretrained latent models, underperform, and demand significant computational resources due to extensive backpropagation through ODE solvers and inversion processes.<n>We propose FlowChef, which leverages the vector field to steer the denoising trajectory for controlled image generation tasks, facilitated by gradient skipping.<n>FlowChef significantly outperforms baselines in terms of performance, memory, and time requirements, achieving new state-of-the
  arXiv  Detail & Related papers  (2024-11-27T19:04:40Z)
• Learning Image Priors through Patch-based Diffusion Models for Solving Inverse Problems [20.72421110904258]
  Diffusion models can learn strong image priors from underlying data distribution and use them to solve inverse problems, but the training process is computationally expensive and requires lots of data.<n>This paper proposes a method to learn an efficient data prior for the entire image by training diffusion models only on patches of images.
  arXiv  Detail & Related papers  (2024-06-04T16:30:37Z)
• Provably Robust Score-Based Diffusion Posterior Sampling for Plug-and-Play Image Reconstruction [31.503662384666274]
  In science and engineering, the goal is to infer an unknown image from a small number of measurements collected from a known forward model describing certain imaging modality. Motivated Score-based diffusion models, due to its empirical success, have emerged as an impressive candidate of an exemplary prior in image reconstruction.
  arXiv  Detail & Related papers  (2024-03-25T15:58:26Z)
• Denoising Diffusion Bridge Models [54.87947768074036]
  Diffusion models are powerful generative models that map noise to data using processes. For many applications such as image editing, the model input comes from a distribution that is not random noise. In our work, we propose Denoising Diffusion Bridge Models (DDBMs)
  arXiv  Detail & Related papers  (2023-09-29T03:24:24Z)
• A Variational Perspective on Solving Inverse Problems with Diffusion Models [101.831766524264]
  Inverse tasks can be formulated as inferring a posterior distribution over data. This is however challenging in diffusion models since the nonlinear and iterative nature of the diffusion process renders the posterior intractable. We propose a variational approach that by design seeks to approximate the true posterior distribution.
  arXiv  Detail & Related papers  (2023-05-07T23:00:47Z)
• Masked Images Are Counterfactual Samples for Robust Fine-tuning [77.82348472169335]
  Fine-tuning deep learning models can lead to a trade-off between in-distribution (ID) performance and out-of-distribution (OOD) robustness. We propose a novel fine-tuning method, which uses masked images as counterfactual samples that help improve the robustness of the fine-tuning model.
  arXiv  Detail & Related papers  (2023-03-06T11:51:28Z)
• Denoising Diffusion Restoration Models [110.1244240726802]
  Denoising Diffusion Restoration Models (DDRM) is an efficient, unsupervised posterior sampling method. We demonstrate DDRM's versatility on several image datasets for super-resolution, deblurring, inpainting, and colorization.
  arXiv  Detail & Related papers  (2022-01-27T20:19:07Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.