Research map / Image generation / Super-resolution
Learning Parametric Sparse Models for Image 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 |
|---|---|---|---|
| Learn sparse priors from external training patches in addition to internal LR-image estimates, versus NCSR (internal only) | sparse-coding SR · Set5 x3 bicubic | PSNR (dB): 33.05 -> 33.39 | Helped |
| Same, Set14 x2 bicubic | sparse-coding SR · Set14 x2 bicubic | PSNR (dB): 32.26 -> 32.61 | Helped |
| Same, BSD100 x4 bicubic | sparse-coding SR · BSD100 x4 bicubic | PSNR (dB): 26.91 -> 27.08 | Helped |
| Proposed versus SRCNN on Set5 x4 bicubic | SRCNN · Set5 x4 bicubic | PSNR (dB): 30.49 -> 31.04 | Helped |
| Proposed versus SRCNN on Set14 x2 bicubic | SRCNN · Set14 x2 bicubic | PSNR (dB): 32.45 -> 32.61 | Helped |
| Proposed versus NCSR on Set5 x2 bicubic SSIM | NCSR · Set5 x2 bicubic | SSIM: 0.9550 -> 0.9551 | No clear effect |
| Proposed versus NCSR on Set14 x2 bicubic SSIM | NCSR · Set14 x2 bicubic | SSIM: 0.9058 -> 0.9072 | No clear effect |
| Learned sparse priors versus NCSR under Gaussian blur, Set5 x3 | NCSR · Set5 x3 Gaussian blur | PSNR (dB): 33.03 -> 33.49 | Helped |
| Learned sparse priors versus NCSR under Gaussian blur, BSD100 x3 | NCSR · BSD100 x3 Gaussian blur | PSNR (dB): 28.35 -> 28.60 | Helped |
| Patch-based example methods (SRCNN) versus internal-prior NCSR on blurred LR images | SRCNN · Set5 x3 Gaussian blur | PSNR (dB): 33.03 -> 30.20 | Hurt |
| Proposed versus NCSR on Set14 x3 blur, SSIM | NCSR · Set14 x3 Gaussian blur | SSIM: 0.8203 -> 0.8255 | Helped |
| Proposed versus SRCNN on image Monarch, blur x3 | SRCNN · Monarch (Set14) x3 blur | PSNR (dB): 29.88 -> 33.84 | Helped |
Closest papers by meaning
- Noise-Free Optimization in Early Training Steps for Image Super-resolution (AAAI 2024 · similarity 0.88)
- Explorable Super Resolution (CVPR 2020 · similarity 0.87)
- Low-Res Leads the Way: Improving Generalization for Super-Resolution by Self-Supervised Learning (CVPR 2024 · similarity 0.87)
- Tackling the Ill-Posedness of Super-Resolution Through Adaptive Target Generation (CVPR 2021 · 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)
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.