Research map / Image generation / Image restoration
Learning Enriched Features for Real Image Restoration and Enhancement
Syed Waqas Zamir, Aditya Arora, Salman Khan, Munawar Hayat, Fahad Shahbaz Khan, Ming-Hsuan Yang, Ling Shao
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 connections (ablation) | MIRNet · RealSR x3 (ablation) | PSNR (dB): 27.91 -> 31.16 | Improved |
| add dual attention units (DAU) (ablation) | MIRNet · RealSR x3 (ablation) | PSNR (dB): 30.97 -> 31.16 | Improved |
| add intermediate SKFF information exchange (ablation) | MIRNet · RealSR x3 (ablation) | PSNR (dB): 30.78 -> 31.16 | Improved |
| add both DAU and intermediate SKFF (ablation) | MIRNet · RealSR x3 (ablation) | PSNR (dB): 30.57 -> 31.16 | Improved |
| use SKFF instead of summation (ablation) | MIRNet · RealSR x3 (ablation) | PSNR (dB): 30.76 -> 31.16 | Improved |
| use SKFF instead of concatenation (ablation) | MIRNet · RealSR x3 (ablation) | PSNR (dB): 30.89 -> 31.16 | Improved |
| increase parallel streams from 1 to 3 at 2 columns (ablation) | MIRNet · RealSR x3 (ablation) | PSNR (dB): 30.11 -> 31.16 | Improved |
| increase columns from 1 to 2 at 3 streams (ablation) | MIRNet · RealSR x3 (ablation) | PSNR (dB): 30.24 -> 31.16 | Improved |
| increase columns from 2 to 3 at 3 streams (ablation) | MIRNet · RealSR x3 (ablation) | PSNR (dB): 31.16 -> 31.18 | No clear effect |
| MIRNet vs VDN on SIDD denoising | MIRNet · SIDD | PSNR (dB): 39.28 -> 39.72 | Improved |
| MIRNet vs VDN on DND denoising (SIDD-trained) | MIRNet · DND (trained on SIDD) | PSNR (dB): 39.38 -> 39.88 | Improved |
| MIRNet vs LP-KPN on RealSR x3 | MIRNet · RealSR x3 | PSNR (dB): 30.42 -> 31.16 | Improved |
Closest papers by meaning
- Learning Enriched Features for Fast Image Restoration and Enhancement (TPAMI 2022 · similarity 0.99)
- Revitalizing Convolutional Network for Image Restoration (TPAMI 2024 · similarity 0.90)
- Omni-Kernel Network for Image Restoration (AAAI 2024 · similarity 0.90)
- Residual Dense Network for Image Restoration (TPAMI 2020 · similarity 0.89)
- Scale-Wise Convolution for Image Restoration (AAAI 2020 · similarity 0.89)
- Unfolding Taylor's Approximations for Image Restoration (NeurIPS 2021 · 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.
