Research map / Image generation / Super-resolution
Image Super-Resolution With Cross-Scale Non-Local Attention and Exhaustive Self-Exemplars Mining
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 CSNLN instead of the strongest listed baseline, SAN | SISR · Set5 ×4 | PSNR: 32.64 -> 32.68 | Helped |
| Use CSNLN instead of the strongest listed baseline, SAN | SISR · Set14 ×4 | PSNR: 28.92 -> 28.95 | Helped |
| Use CSNLN instead of the strongest listed baseline, SAN | SISR · B100 ×4 | PSNR: 27.78 -> 27.80 | Helped |
| Use CSNLN instead of the strongest listed baseline, RCAN | SISR · Urban100 ×4 | PSNR: 26.82 -> 27.22 | Helped |
| Use CSNLN instead of the strongest listed baseline, RCAN | SISR · Manga109 ×4 | PSNR: 31.22 -> 31.43 | Helped |
| Add the local branch to the branch-free recurrent baseline (ablation) | SEM · Set14 ×2 | PSNR: 33.32 -> 33.47 | Helped |
| Add the in-scale non-local branch to the branch-free recurrent baseline (ablation) | SEM · Set14 ×2 | PSNR: 33.32 -> 33.52 | Helped |
| Add the cross-scale non-local branch to the branch-free recurrent baseline (ablation) | SEM · Set14 ×2 | PSNR: 33.32 -> 33.51 | Helped |
| Combine local, in-scale non-local, and cross-scale non-local branches (ablation) | SEM · Set14 ×2 | PSNR: 33.62 -> 33.74 | Helped |
| Use 3×3 rather than 1×1 matching patches (ablation) | CS-NL attention · Set14 ×2 | PSNR: 33.67 -> 33.74 | Helped |
| Increase matching patch size from 3×3 to 5×5 (ablation) | CS-NL attention · Set14 ×2 | PSNR: 33.74 -> 33.61 | Hurt |
| Use mutual-projected fusion instead of addition (ablation) | SEM · Set14 ×2 | PSNR: 33.69 -> 33.74 | Helped |
Closest papers by meaning
- Non-Local Recurrent Network for Image Restoration (NeurIPS 2018 · similarity 0.90)
- Second-Order Attention Network for Single Image Super-Resolution (CVPR 2019 · similarity 0.90)
- Image Super-Resolution With Non-Local Sparse Attention (CVPR 2021 · similarity 0.90)
- Cross-Scale Internal Graph Neural Network for Image Super-Resolution (NeurIPS 2020 · similarity 0.89)
- SAVSR: Arbitrary-Scale Video Super-Resolution via a Learned Scale-Adaptive Network (AAAI 2024 · similarity 0.88)
- Transcending the Limit of Local Window: Advanced Super-Resolution Transformer with Adaptive Token Dictionary (CVPR 2024 · similarity 0.87)
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.