Research map / Image generation / Super-resolution
Image Super-Resolution via Deep Recursive Residual Network
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 |
|---|---|---|---|
| Use DRRN B1U25 rather than the strongest prior method in the table | SISR models · Urban100 ×2 | PSNR: +0.47 dB | Helped |
| Use DRRN B1U25 rather than the strongest prior method in the table | SISR models · Urban100 ×3 | PSNR: +0.38 dB | Helped |
| Use DRRN B1U25 rather than the strongest prior method in the table | SISR models · Urban100 ×4 | PSNR: +0.26 dB | Helped |
| Add local residual learning (ablation) | VDSR-style SISR network · Set5 ×3 | PSNR: 33.86 -> 33.92 | Helped |
| Add multi-path structure (ablation) | DRRN chain-structure variant · Set5 ×3 | PSNR: 33.95 -> 33.99 | Helped |
| Add recursive weight sharing (ablation) | DRRN without shared residual-unit weights · Set5 ×3 | PSNR: 33.97 -> 33.99 | Helped |
| Increase depth from 20 to 52 layers without adding parameters (ablation) | DRRN B1U9 · Set5 ×3 | PSNR: 33.93 -> 34.03 | Helped |
| Vary recursive block count and units per block while increasing depth (ablation) | DRRN · Set5 ×3 | PSNR | Helped |
| Use recursive learning in B1U25 rather than no recursive learning in B17U1 (ablation) | DRRN, depth 52 · Set5 ×3 | PSNR: B17U1 34.03 dB; B1U25 34.03 dB | No clear effect |
| Replace chain-mode residual units with multi-path residual units (ablation) | DRRN · Set5 ×3 | PSNR | Helped |
Closest papers by meaning
- Deeply-Recursive Convolutional Network for Image Super-Resolution (CVPR 2016 · similarity 0.94)
- Image Super-Resolution via Dual-State Recurrent Networks (CVPR 2018 · similarity 0.89)
- Accurate Image Super-Resolution Using Very Deep Convolutional Networks (CVPR 2016 · similarity 0.89)
- Post-trained Convolution Networks for Single Image Super-resolution (Abstract Reprint) (AAAI 2024 · similarity 0.88)
- Feedback Network for Image Super-Resolution (CVPR 2019 · similarity 0.88)
- ResDiff: Combining CNN and Diffusion Model for Image Super-resolution (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.