Research map / Image generation / Super-resolution
Unfolding the Alternating Optimization for Blind Super Resolution
Read the paperCheck 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 |
|---|---|---|---|
| use end-to-end alternating Estimator–Restorer optimization instead of the two-step KernelGAN + ZSSR pipeline | Blind SR · DIV2KRK ×2 | PSNR: 30.36 -> 32.56 | Helped |
| use end-to-end alternating Estimator–Restorer optimization instead of the two-step KernelGAN + ZSSR pipeline | Blind SR · DIV2KRK ×4 | PSNR: 26.81 -> 27.55 | Helped |
| replace the two-step IKC solution with DAN | Blind SR · Manga109 Gaussian8 ×3 | PSNR: 28.21 -> 33.16 | Helped |
| replace the two-step IKC solution with DAN | Blind SR · Set5, Set14, BSD100, Urban100, Manga109 Gaussian8 ×2–×4 | PSNR | Helped |
| replace estimated kernels with ground-truth kernels for the Restorer (ablation) | DAN · Set5, Set14, BSD100, Urban100, Manga109 Gaussian8 ×4 | PSNR: 31.89 -> 31.85; 28.43 -> 28.42; 27.51 -> 27.51; 25.86 -> 25.87; 30.50 -> 30.51 | No clear effect |
| test DAN with more iterations than its four-iteration training configuration (ablation) | DAN · Set14 | PSNR: 20.42dB -> 20.43dB | No clear effect |
| use DAN instead of KernelGAN + ZSSR | Blind SR · Gaussian8 images from Set5 | inference time per image: 415.7 seconds -> 0.75 seconds | Helped |
| use DAN instead of IKC | Blind SR · Gaussian8 images from Set5 | inference time per image: 3.93 seconds -> 0.75 seconds | Helped |
Closest papers by meaning
- Blind Super-Resolution With Iterative Kernel Correction (CVPR 2019 · similarity 0.92)
- Deep Constrained Least Squares for Blind Image Super-Resolution (CVPR 2022 · similarity 0.91)
- A Dynamic Kernel Prior Model for Unsupervised Blind Image Super-Resolution (CVPR 2024 · similarity 0.90)
- Noise-Free Optimization in Early Training Steps for Image Super-resolution (AAAI 2024 · similarity 0.88)
- Unsupervised Degradation Representation Aware Transform for Real-World Blind Image Super-Resolution (AAAI 2025 · similarity 0.87)
- Navigating Beyond Dropout: An Intriguing Solution towards Generalizable Image Super Resolution (CVPR 2024 · 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.