Research map / Image generation / Super-resolution
Image Super-Resolution Using Dense Skip Connections
Tong Tong, Gen Li, Xiejie Liu, Qinquan Gao
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 skip connection combining low-level and high-level features (ablation) | SRDenseNet H · Urban100 x4 | PSNR: 25.69 -> 25.86 | Improved |
| add dense skip connections from all levels of features (ablation) | SRDenseNet HL · Urban100 x4 | PSNR: 25.86 -> 26.05 | Improved |
| add skip connection combining low-level and high-level features (ablation) | SRDenseNet H · Set5 x4 | PSNR: 31.66 -> 31.80 | Improved |
| add dense skip connections from all levels of features (ablation) | SRDenseNet HL · Set5 x4 | PSNR: 31.80 -> 32.02 | Improved |
| add skip connection combining low-level and high-level features (ablation) | SRDenseNet H · Set14 x4 | PSNR: 28.34 -> 28.40 | Improved |
| add dense skip connections from all levels of features (ablation) | SRDenseNet HL · Set14 x4 | PSNR: 28.40 -> 28.50 | Improved |
| add skip connection combining low-level and high-level features (ablation) | SRDenseNet H · B100 x4 | PSNR: 27.42 -> 27.47 | Improved |
| add dense skip connections from all levels of features (ablation) | SRDenseNet HL · B100 x4 | PSNR: 27.47 -> 27.53 | Improved |
| use SRDenseNet All instead of VDSR (20-layer CNN) | VDSR · Urban100 x4 | PSNR: 25.18 -> 26.05 | Improved |
| use SRDenseNet All instead of DRCN | DRCN · Set5 x4 | PSNR: 31.53 -> 32.02 | Improved |
| use SRDenseNet All instead of SRCNN | SRCNN · All (219 images) x4 | PSNR: 25.93 -> 27.02 | Improved |
Closest papers by meaning
- Image Super-Resolution Using Deep Convolutional Networks (TPAMI 2016 · similarity 0.91)
- Accelerating the Super-Resolution Convolutional Neural Network (ECCV 2016 · similarity 0.90)
- Accurate Image Super-Resolution Using Very Deep Convolutional Networks (CVPR 2016 · similarity 0.90)
- Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network (CVPR 2016 · similarity 0.90)
- Deep Back-Projection Networks for Super-Resolution (CVPR 2018 · similarity 0.90)
- Image Super-Resolution Using Very Deep Residual Channel Attention Networks (ECCV 2018 · similarity 0.89)
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.
