Research map / Image generation / Super-resolution
Global Learnable Attention for Single Image Super-Resolution
Jian-Nan Su, Min Gan, Guangyong Chen, Jia-Li Yin, C. L. Philip Chen
AI-extracted. The measured effects below were read by AI from the paper's full text (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 learnable similarity scoring s_l to fixed dot product in GLA (no LFFB) (ablation) | DLSN · Set14 (x2) | PSNR (dB): 33.33 -> 33.49 | Improved |
| add learnable similarity scoring s_l on top of fixed dot product (with LFFB) (ablation) | DLSN · Set14 (x2) | PSNR (dB): 33.65 -> 33.77 | Improved |
| add LFFB local features fusion block (ablation) | DLSN · Set14 (x2) | PSNR (dB): 33.49 -> 33.77 | Improved |
| add GLA to LFFB (ablation) | DLSN · Set14 (x2) | PSNR (dB): 33.54 -> 33.77 | Improved |
| increase hashing rounds in training and testing from h=1 to h=4 (ablation) | DLSN · Set14 (x2) | PSNR (dB): 33.60 -> 33.78 | Improved |
| use h=3 instead of h=4 hashing rounds (ablation) | DLSN · Set14 (x2) | PSNR (dB): 33.77 -> 33.78 | No clear effect |
| enlarge bucket size beyond l=128 (ablation) | DLSN · Set14 (x2) | PSNR (dB) | Hurt |
| integrate GLA into FSRCNN | FSRCNN · SISR benchmarks | PSNR (dB): +0.13dB | Improved |
| integrate GLA into EDSR | EDSR · SISR benchmarks | PSNR (dB): +0.14dB | Improved |
| integrate GLA into RCAN | RCAN · SISR benchmarks | PSNR (dB): +0.17dB | Improved |
| DLSN vs NLSN at x3 on Urban100 | DLSN · Urban100 (x3) | PSNR (dB): +0.29dB | Improved |
| DLSN (C=128) vs NLSN with fewer parameters | DLSN · Manga109 (x4) | PSNR (dB): +0.38dB | Improved |
Closest papers by meaning
- Revealing the Dark Side of Non-Local Attention in Single Image Super-Resolution (TPAMI 2024 · similarity 0.91)
- Image Super-Resolution With Non-Local Sparse Attention (CVPR 2021 · similarity 0.89)
- Learning Texture Transformer Network for Image Super-Resolution (CVPR 2020 · similarity 0.88)
- Image Super-Resolution With Cross-Scale Non-Local Attention and Exhaustive Self-Exemplars Mining (CVPR 2020 · similarity 0.87)
- Recursive Generalization Transformer for Image Super-Resolution (ICLR 2024 · similarity 0.86)
- SeD: Semantic-Aware Discriminator for Image Super-Resolution (CVPR 2024 · similarity 0.85)
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.
