Research map / Image generation / Super-resolution
Closed-Loop Matters: Dual Regression Networks for Single Image Super-Resolution
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 dual regression to reconstruct the input LR image (ablation) | DRN-S · Set5, 4× SR | PSNR: 32.53 -> 32.68 | Helped |
| Add dual regression to reconstruct the input LR image (ablation) | DRN-L · Set5, 4× SR | PSNR: 32.61 -> 32.74 | Helped |
| Set λ to 0.1 rather than 0.001 (ablation) | DRN-S · Set5, 4× SR | PSNR: 32.57 -> 32.67 | Helped |
| Increase λ from 0.1 to 1.0 (ablation) | DRN-S · Set5, 4× SR | PSNR: 32.67 -> 32.51 | Hurt |
| Increase λ from 0.1 to 10 (ablation) | DRN-S · Set5, 4× SR | PSNR: 32.67 -> 32.37 | Hurt |
| Add dual regression on HR images as well as LR images (ablation) | DRN-S · Set5, 4× SR | PSNR: 32.68 -> 32.69 | No clear effect |
| Replace Bicubic with Nearest degradation for paired synthetic data (ablation) | DRN-Adapt · Set14, 8× SR, BD degradation | PSNR: 23.07 -> 23.03 | No clear effect |
| Add dual regression to SRResNet (ablation) | SRResNet · Set5, 4× SR | PSNR: 32.26 -> 32.47 | Helped |
| Adapt using dual regression and unpaired LR data alongside paired synthetic data | DRN-S · Set5, 8× SR, Nearest degradation | PSNR: 21.22 -> 23.00 | Helped |
| Adapt using dual regression and unpaired LR data alongside paired synthetic data | DRN-S · Set5, 8× SR, BD degradation | PSNR: 23.54 -> 24.62 | Helped |
| Use DRN-L rather than RRDB | 4× SR · Set5 | PSNR: 32.73 -> 32.74 | Helped |
| Use DRN-L rather than RCAN | 8× SR · Manga109 | PSNR: 25.23 -> 25.33 | Helped |
Closest papers by meaning
- Learning Controllable Degradation for Real-World Super-Resolution via Constrained Flows (ICML 2023 · similarity 0.90)
- Correction Filter for Single Image Super-Resolution: Robustifying Off-the-Shelf Deep Super-Resolvers (CVPR 2020 · similarity 0.90)
- Low-Res Leads the Way: Improving Generalization for Super-Resolution by Self-Supervised Learning (CVPR 2024 · similarity 0.89)
- Noise-Free Optimization in Early Training Steps for Image Super-resolution (AAAI 2024 · similarity 0.89)
- Learning Coupled Dictionaries from Unpaired Data for Image Super-Resolution (CVPR 2024 · similarity 0.88)
- SAVSR: Arbitrary-Scale Video Super-Resolution via a Learned Scale-Adaptive Network (AAAI 2024 · similarity 0.88)
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.