Research map / Image generation / Super-resolution
Densely Residual Laplacian Super-Resolution
Saeed Anwar, Nick M. Barnes
Read the paperCheck your idea against 137,217 papers
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 Laplacian attention (ablation) | DRLN with DC, MSC, LSC · Set5 (2x bicubic) | PSNR (dB): 32.30 -> 32.37 | Improved |
| add dense connections (ablation) | MSC+LSC+LA network · Set5 (2x bicubic) | PSNR (dB): 32.06 -> 32.37 | Improved |
| add medium skip connections (ablation) | DC+LSC+LA network · Set5 (2x bicubic) | PSNR (dB): 32.12 -> 32.37 | Improved |
| add long skip connection (ablation) | DC+MSC network · Set5 (2x bicubic) | PSNR (dB): 32.10 -> 32.37 | Improved |
| add all connection types and attention (ablation) | plain deep network · Set5 (2x bicubic) | PSNR (dB): 31.92 -> 32.37 | Improved |
| add Laplacian attention (ablation) | MSC+LSC network · Set5 (2x bicubic) | PSNR (dB): 32.06 -> 32.06 | No clear effect |
| add Laplacian attention (ablation) | DC+LSC network · Set5 (2x bicubic) | PSNR (dB): 31.85 -> 32.12 | Improved |
| add self-ensemble (DRLN+) | DRLN · Urban100 (4x bicubic) | PSNR (dB): 26.98 -> 27.14 | Improved |
| use DRLN instead of RCAN as SR front end | ResNet-50 classifier · ImageNet 1k images | Top-1 error: 0.393 -> 0.345 | Improved |
| use DRLN instead of RCAN | 4x SR · MANGA109 | runtime (s): 1.14s -> 0.045s | Improved |
| use DRLN instead of RCAN | 3x blur-down SR · Urban100 | PSNR (dB): 28.81 -> 29.11 | Improved |
Closest papers by meaning
- Deep Laplacian Pyramid Networks for Fast and Accurate Super-Resolution (CVPR 2017 · similarity 0.94)
- Fast and Accurate Image Super-Resolution with Deep Laplacian Pyramid Networks (TPAMI 2018 · similarity 0.92)
- Fast, Accurate, and Lightweight Super-Resolution with Cascading Residual Network (ECCV 2018 · similarity 0.90)
- Multi-scale Residual Network for Image Super-Resolution (ECCV 2018 · similarity 0.89)
- Residual Dense Network for Image Super-Resolution (CVPR 2018 · similarity 0.88)
- ResDiff: Combining CNN and Diffusion Model for Image Super-resolution (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.
