Research map / Image generation / Super-resolution
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
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add superpixel segmentation module (ablation) | CASR base model · DIV8K (x4 x3 x1.5 to x18) | MUSIQ: 31.73 -> 42.23 | Improved |
| add superpixel segmentation module (ablation) | CASR base model · DIV8K (x4 x3 x1.5 to x18) | PI: 5.91 -> 5.96 | No clear effect |
| add depth conditioning to superpixel model (ablation) | CASR · DIV8K (x4 x3 x1.5 to x18) | MUSIQ: 42.23 -> 45.18 | Improved |
| add SARM to superpixel+depth model (ablation) | CASR · DIV8K (x4 x3 x1.5 to x18) | MUSIQ: 45.18 -> 51.44 | Improved |
| add cross-attention with LR global semantics to SARM (ablation) | CASR SARM · DIV8K (x4 x3 x1.5 to x18) | MUSIQ: 49.10 -> 51.44 | Improved |
| add cross-attention with LR global semantics to SARM (ablation) | CASR SARM · DIV8K (x4 x3 x1.5 to x18) | NIQE: 6.30 -> 6.01 | Improved |
| add correlation loss L_corr (ablation) | CASR · DIV8K (x4 x3 x1.5 to x18) | MUSIQ: 49.33 -> 50.23 | Improved |
| add correlation loss L_corr (ablation) | CASR · DIV8K (x4 x3 x1.5 to x18) | NIQE: 6.71 -> 6.24 | Improved |
| add depth loss L_depth (ablation) | CASR · DIV8K (x4 x3 x1.5 to x18) | LPIPS: 0.459 -> 0.450 | Improved |
| increase superpixel size from 4x4 to 8x8 (ablation) | CASR · DIV8K (x18) | MUSIQ: 51.44 -> 64.29 | Mixed |
| reduce superpixel size from 4x4 to 3x3 (ablation) | CASR · DIV8K (x18) | LPIPS: 0.450 -> 0.513 | Hurt |
| increase superpixel size from 4x4 to 5x5 (ablation) | CASR · DIV8K (x18) | LPIPS: 0.450 -> 0.481 | Hurt |
Later papers that compared with this method
1 later accepted paper reports CASR as a baseline in their results tables; 0 of them report a better number on most of the results they share with it.
- MFmamba: A Multi-function Network for Panchromatic Image Resolution Restoration Based on State-Space Model (AAAI 2026 · better on 1 of 2 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- StructSR: Refuse Spurious Details in Real-World Image Super-Resolution (AAAI 2025 · similarity 0.87)
- Chain-of-Zoom: Extreme Super-Resolution via Scale Autoregression and Preference Alignment (NeurIPS 2025 · similarity 0.87)
- BUFF: Bayesian Uncertainty Guided Diffusion Probabilistic Model for Single Image Super-Resolution (AAAI 2025 · similarity 0.87)
- SinSR: Diffusion-Based Image Super-Resolution in a Single Step (CVPR 2024 · similarity 0.86)
- Pixel-level and Semantic-level Adjustable Super-resolution: A Dual-LoRA Approach (CVPR 2025 · similarity 0.86)
- NSARM: Next-Scale Autoregressive Modeling for Robust Real-World Image Super-Resolution (NeurIPS 2026 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
