Research map / Image generation / Super-resolution
Blind Super-Resolution With Iterative Kernel Correction
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 iterative kernel correction to predictor-plus-SFTMD SR | SFTMD blind SR · Set5, Set14, BSD100, Urban100, Manga109 (×2, ×3, ×4) | PSNR: 35.44 -> 36.62 (Set5 ×2); 31.27 -> 32.82 (Set14 ×2); 30.54 -> 31.36 (BSD100 ×2); 27.80 -> 30.36 (Urban100 ×2); 30.75 -> | Helped |
| Use SFT conditioning rather than direct feature concatenation (ablation) | SRResNet-based multi-kernel SR · Set5, kernel width 0.2, ×2 | PSNR: 37.69 -> 38.00 | Helped |
| Remove PCA kernel representation (ablation) | IKC · BSD100, kernel width 2.0 and 3.0 (two settings) | PSNR: 26.85 -> 27.06 (kernel width 2.0, first setting); 26.75 -> 26.98 (kernel width 3.0, first setting) | Hurt |
Closest papers by meaning
- Spectrum-to-Kernel Translation for Accurate Blind Image Super-Resolution (NeurIPS 2021 · similarity 0.93)
- Deep Constrained Least Squares for Blind Image Super-Resolution (CVPR 2022 · similarity 0.92)
- Unfolding the Alternating Optimization for Blind Super Resolution (NeurIPS 2020 · similarity 0.92)
- A Dynamic Kernel Prior Model for Unsupervised Blind Image Super-Resolution (CVPR 2024 · similarity 0.89)
- Navigating Beyond Dropout: An Intriguing Solution towards Generalizable Image Super Resolution (CVPR 2024 · similarity 0.86)
- Unsupervised Degradation Representation Aware Transform for Real-World Blind Image Super-Resolution (AAAI 2025 · similarity 0.85)
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.