Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

CASR: A Robust Cyclic Framework for Arbitrary Large-Scale Super-Resolution with Distribution Alignment and Self-Similarity Awareness

Wenhao Guo, Zhaoran Zhao, Peng Lu, Sheng Li, Qian Qiao, RuiDe Li

TL;DR

This work proposes CASR, a simple yet highly efficient cyclic SR framework that reformulates ultra-magnification as a sequence of in-distribution scale transitions that ensures stable inference at arbitrary scales while requiring only a single model.

Abstract

Arbitrary-Scale SR (ASISR) remains fundamentally limited by cross-scale distribution shift: once the inference scale leaves the training range, noise, blur, and artifacts accumulate sharply. We revisit this challenge from a cross-scale distribution transition perspective and propose CASR, a simple yet highly efficient cyclic SR framework that reformulates ultra-magnification as a sequence of in-distribution scale transitions. This design ensures stable inference at arbitrary scales while requiring only a single model. CASR tackles two major bottlenecks: distribution drift across iterations and patch-wise diffusion inconsistencies. The proposed SDAM module aligns structural distributions via superpixel aggregation, preventing error accumulation, while SARM module restores high-frequency textures by enforcing autocorrelation and embedding LR self-similarity priors. Despite using only a single model, our approach significantly reduces distribution drift, preserves long-range texture consistency, and achieves superior generalization even at extreme magnification.

CASR: A Robust Cyclic Framework for Arbitrary Large-Scale Super-Resolution with Distribution Alignment and Self-Similarity Awareness

TL;DR

This work proposes CASR, a simple yet highly efficient cyclic SR framework that reformulates ultra-magnification as a sequence of in-distribution scale transitions that ensures stable inference at arbitrary scales while requiring only a single model.

Abstract

Arbitrary-Scale SR (ASISR) remains fundamentally limited by cross-scale distribution shift: once the inference scale leaves the training range, noise, blur, and artifacts accumulate sharply. We revisit this challenge from a cross-scale distribution transition perspective and propose CASR, a simple yet highly efficient cyclic SR framework that reformulates ultra-magnification as a sequence of in-distribution scale transitions. This design ensures stable inference at arbitrary scales while requiring only a single model. CASR tackles two major bottlenecks: distribution drift across iterations and patch-wise diffusion inconsistencies. The proposed SDAM module aligns structural distributions via superpixel aggregation, preventing error accumulation, while SARM module restores high-frequency textures by enforcing autocorrelation and embedding LR self-similarity priors. Despite using only a single model, our approach significantly reduces distribution drift, preserves long-range texture consistency, and achieves superior generalization even at extreme magnification.
Paper Structure (20 sections, 7 equations, 10 figures, 6 tables)

This paper contains 20 sections, 7 equations, 10 figures, 6 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Arbitrary-Scale Super-Resolution
  4. Large-scale Image Super-Resolution
  5. Method
  6. Superpixel-based Distribution Alignment
  7. Self-Similarity Aware Refinement
  8. Training Strategy
  9. Experiments
  10. Training Datasets and Metrics
  11. Testing Datasets
  12. Implementation Details
  13. Comparisons with State-of-the-Art
  14. Experiments on Synthetic Dataset
  15. Experiments on Real-World Datasets
  16. ...and 5 more sections

Figures (10)

  • Figure 1: Comparison of cyclic cascade stability across different ASISR. The SIFID measures distribution shifts between reconstructed images and the training data during cascading. Our method achieves notably higher distribution stability than others.
  • Figure 2: This illustrates the texture inconsistency between patches caused by patch-based super-resolution, where identical repeated objects are reconstructed with different texture patterns.
  • Figure 3: Illustration of the proposed CASR. The purple module denotes the SDAM, the green block corresponds to the SARM, and the gray U-Net represents the SR backbone.
  • Figure 4: Illustration of the distribution alignment process, where the input image is decomposed into a superpixel representation and a depth map. This decomposition effectively removes artifacts and noise, enabling robust SR.
  • Figure 5: Illustration of the local self-similarity computation, where structurally similar regions are assigned higher correlation.
  • ...and 5 more figures