Pith. sign in

REVIEW 9 cited by

Inversion by Direct Iteration: An Alternative to Denoising Diffusion for Image Restoration

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2303.11435 v5 pith:I5B76VGU submitted 2023-03-20 eess.IV cs.CVcs.LG

Mauricio Delbracio , Peyman Milanfar This is my paper
classification eess.IVcs.CVcs.LG
keywords imagedenoisingrestorationdiffusionindihigh-qualityinputlow-quality
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Inversion by Direct Iteration (InDI) is a new formulation for supervised image restoration that avoids the so-called "regression to the mean" effect and produces more realistic and detailed images than existing regression-based methods. It does this by gradually improving image quality in small steps, similar to generative denoising diffusion models. Image restoration is an ill-posed problem where multiple high-quality images are plausible reconstructions of a given low-quality input. Therefore, the outcome of a single step regression model is typically an aggregate of all possible explanations, therefore lacking details and realism. The main advantage of InDI is that it does not try to predict the clean target image in a single step but instead gradually improves the image in small steps, resulting in better perceptual quality. While generative denoising diffusion models also work in small steps, our formulation is distinct in that it does not require knowledge of any analytic form of the degradation process. Instead, we directly learn an iterative restoration process from low-quality and high-quality paired examples. InDI can be applied to virtually any image degradation, given paired training data. In conditional denoising diffusion image restoration the denoising network generates the restored image by repeatedly denoising an initial image of pure noise, conditioned on the degraded input. Contrary to conditional denoising formulations, InDI directly proceeds by iteratively restoring the input low-quality image, producing high-quality results on a variety of image restoration tasks, including motion and out-of-focus deblurring, super-resolution, compression artifact removal, and denoising.

Discussion (0). Sign in to comment.

Forward citations

Cited by 9 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. DIPA: Distilled Preconditioned Algorithms for Solving Imaging Inverse Problems

    eess.IV 2026-05 unverdicted novelty 7.0 of 10

    DIPA learns preconditioning operators via distillation from a teacher with a better sensing matrix to improve reconstruction quality for the student's physically constrained matrix in imaging inverse problems.

  2. ScaleResfusion: Residual Rectified Flow based on Residual Vector Field

    cs.CV 2026-07 conditional novelty 6.0 of 10

    ScaleResfusion modifies rectified flow to start from a noisy low-quality image and learn only a residual velocity field, enabling 4-step image restoration with LoRA fine-tuning of billion-scale text-to-image models.

  3. Stop Denoising Your Blurs

    cs.CV 2026-05 unverdicted novelty 6.0 of 10

    ConvDiff builds a convolution-based diffusion trajectory from clean to blurred images in the frequency domain for Gaussian deblurring.

  4. A Noise Constrained Diffusion (NC-Diffusion) Framework for High Fidelity Image Compression

    eess.IV 2026-04 unverdicted novelty 6.0 of 10

    NC-Diffusion matches quantization noise to the diffusion forward process, adds an adaptive frequency filter and zero-shot enhancement, and reports superior fidelity on benchmarks.

  5. Deep Distillation Gradient Preconditioning for Inverse Problems

    eess.IV 2025-08 unverdicted novelty 6.0 of 10

    Knowledge distillation trains a nonlinear gradient preconditioner that improves convergence and reconstruction quality for plug-and-play FISTA on ill-conditioned inverse imaging problems.

  6. Unifying Deep Stochastic Processes for Image Enhancement

    cs.CV 2026-05 unverdicted novelty 5.0 of 10

    Stochastic image enhancement methods are shown to be variants of a shared SDE differing in drift, diffusion, terminal distributions and boundary conditions, with controlled experiments revealing no single dominant fam...

  7. Training-inference input alignment outweighs framework choice in longitudinal retinal image prediction

    cs.CV 2026-04 unverdicted novelty 5.0 of 10

    Training-inference input alignment outweighs framework choice for longitudinal retinal image prediction, with deterministic regression matching complex models when acquisition variability dominates disease progression.

  8. SPROUT: A Scalable Diffusion Foundation Model for Agricultural Vision

    cs.CV 2026-03 conditional novelty 5.0 of 10

    SPROUT, a pixel-space diffusion transformer pre-trained on 2.6M unlabeled agricultural images with effective-rank timestep selection, outperforms web-pretrained foundation models on dense plant phenotyping tasks.

  9. On Inverse Problems, Parameter Estimation, and Domain Generalization

    cs.IT 2025-06 unverdicted novelty 5.0 of 10

    A theoretical framework for parameter estimation in inverse problems shows inversion does not necessarily improve accuracy per the data processing inequality and reveals a vulnerability in domain generalization via th...

Pith tools