Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

UHD Image Deblurring via Autoregressive Flow with Ill-conditioned Constraints

Yucheng Xin, Dawei Zhao, Xiang Chen, Chen Wu, Pu Wang, Dianjie Lu, Guijuan Zhang, Xiuyi Jia, Zhuoran Zheng

TL;DR

The core idea is to decompose UHD restoration into a progressive, coarse-to-fine process: at each scale, the sharp estimate is formed by upsampling the previous-scale result and adding a current-scale residual, enabling stable, stage-wise refinement from low to high resolution.

Abstract

Ultra-high-definition (UHD) image deblurring poses significant challenges for UHD restoration methods, which must balance fine-grained detail recovery and practical inference efficiency. Although prominent discriminative and generative methods have achieved remarkable results, a trade-off persists between computational cost and the ability to generate fine-grained detail for UHD image deblurring tasks. To further alleviate these issues, we propose a novel autoregressive flow method for UHD image deblurring with an ill-conditioned constraint. Our core idea is to decompose UHD restoration into a progressive, coarse-to-fine process: at each scale, the sharp estimate is formed by upsampling the previous-scale result and adding a current-scale residual, enabling stable, stage-wise refinement from low to high resolution. We further introduce Flow Matching to model residual generation as a conditional vector field and perform few-step ODE sampling with efficient Euler/Heun solvers, enriching details while keeping inference affordable. Since multi-step generation at UHD can be numerically unstable, we propose an ill-conditioning suppression scheme by imposing condition-number regularization on a feature-induced attention matrix, improving convergence and cross-scale consistency. Our method demonstrates promising performance on blurred images at 4K (3840$\times$2160) or higher resolutions.

UHD Image Deblurring via Autoregressive Flow with Ill-conditioned Constraints

TL;DR

The core idea is to decompose UHD restoration into a progressive, coarse-to-fine process: at each scale, the sharp estimate is formed by upsampling the previous-scale result and adding a current-scale residual, enabling stable, stage-wise refinement from low to high resolution.

Abstract

Ultra-high-definition (UHD) image deblurring poses significant challenges for UHD restoration methods, which must balance fine-grained detail recovery and practical inference efficiency. Although prominent discriminative and generative methods have achieved remarkable results, a trade-off persists between computational cost and the ability to generate fine-grained detail for UHD image deblurring tasks. To further alleviate these issues, we propose a novel autoregressive flow method for UHD image deblurring with an ill-conditioned constraint. Our core idea is to decompose UHD restoration into a progressive, coarse-to-fine process: at each scale, the sharp estimate is formed by upsampling the previous-scale result and adding a current-scale residual, enabling stable, stage-wise refinement from low to high resolution. We further introduce Flow Matching to model residual generation as a conditional vector field and perform few-step ODE sampling with efficient Euler/Heun solvers, enriching details while keeping inference affordable. Since multi-step generation at UHD can be numerically unstable, we propose an ill-conditioning suppression scheme by imposing condition-number regularization on a feature-induced attention matrix, improving convergence and cross-scale consistency. Our method demonstrates promising performance on blurred images at 4K (38402160) or higher resolutions.
Paper Structure (31 sections, 20 equations, 5 figures, 2 tables)

This paper contains 31 sections, 20 equations, 5 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Methodology
  4. Overview
  5. Autoregressive Architecture
  6. Scale Sequence Construction
  7. Autoregressive Residual Fusion
  8. Condition Injection Mechanism
  9. Rectified Flow Generative Modeling
  10. Rectified Flow Basics
  11. Flow Matching
  12. ODE Solving
  13. Condition-Number Regularization
  14. Attention Matrix Construction
  15. Condition Number Constraint
  16. ...and 16 more sections

Figures (5)

  • Figure 1: Runtime vs. Quality trade-off on UHD deblurring benchmarks. We plot PSNR against per-image runtime (log-scale, in milliseconds) for representative baselines and ARF-IC (Ours) on UHD-Blur and MC-Blur (UHDM).
  • Figure 2: The overall framework of the proposed UHD deblurring method. We begin with coarse-to-fine reconstruction using rectified residual sampling, stabilizing the model's regression capability by constraining the ill-conditionedness of features.
  • Figure 3: Qualitative comparison on UHD deblurring benchmarks. From top to bottom, the first and second rows are UHD-Blur, and the third and fourth rows are MC-Blur(UHDM).
  • Figure 4: Qualitative comparison on non-UHD deblurring benchmarks. From top to bottom, the rows correspond to GoPro, DVD, RealBlur-R, and RealBlur-J.
  • Figure 5: Ablation study visualizations. Top row: PSNR--time trade-offs for settings affecting runtime. Bottom row: PSNR-only ablations.