Research map / Image generation / Image restoration
Residual Non-local Attention Networks for Image Restoration
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 |
|---|---|---|---|
| Enable non-local mixed attention instead of removing all mask branches and non-local blocks (ablation) | RNAN · Urban100 denoising σ=30 | PSNR (dB): 30.96 -> 31.32 | Helped |
| Add a mask branch (ablation) | RNAN · Urban100 denoising σ=30 | PSNR (dB): 30.96 -> 31.17 | Helped |
| Add non-local blocks (ablation) | RNAN · Urban100 denoising σ=30 | PSNR (dB): 30.96 -> 31.20 | Helped |
| Increase non-local block count (ablation) | RNAN · Urban100 denoising σ=30 | PSNR (dB): 31.17 -> 31.32 | Helped |
| Increase local attention block count while fixing RNAB count (ablation) | RNAN · Urban100 denoising σ=30 | PSNR (dB): 30.78 -> 31.27; 30.99 -> 31.50 | Helped |
| Use RNAN rather than FFDNet | Color denoising · Urban100 σ=70 | PSNR (dB): 26.39 -> 27.45 | Helped |
| Use RNAN rather than IRCNN | Image demosaicing · Kodak24 | PSNR: 40.41 -> 43.09 | Helped |
| Use RNAN rather than DnCNN | Compression artifacts reduction · LIVE1 q=10 | PSNR: 29.19 -> 29.63 | Helped |
| Add self-ensemble | RNAN · Super-resolution ×4, Set14 | PSNR: 28.83 -> 28.90 | Helped |
| Add self-ensemble | RNAN · Super-resolution ×4, Urban100 | PSNR: 26.61 -> 26.75 | Helped |
| Replace RCAN with self-ensembled RNAN | Super-resolution ×4 · Urban100 | PSNR: 26.82 -> 26.75 | Hurt |
Closest papers by meaning
- Non-Local Recurrent Network for Image Restoration (NeurIPS 2018 · similarity 0.89)
- KNN Local Attention for Image Restoration (CVPR 2022 · similarity 0.89)
- Image Restoration Using Very Deep Convolutional Encoder-Decoder Networks with Symmetric Skip Connections (NeurIPS 2016 · similarity 0.87)
- Dual Residual Networks Leveraging the Potential of Paired Operations for Image Restoration (CVPR 2019 · similarity 0.87)
- ACL: Activating Capability of Linear Attention for Image Restoration (CVPR 2025 · similarity 0.86)
- Omni-Kernel Network for Image Restoration (AAAI 2024 · 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.