Research map / Image generation / Super-resolution
End-to-End Learning for Joint Image Demosaicing, Denoising and Super-Resolution
Check 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 |
|---|---|---|---|
| Use JDNDMSR+ instead of TENet | Joint denoising, demosaicing, and ×2 super-resolution · McMaster, Kodak, B100, and Urban100 at σ=0, 10, 20, and 30 | cPSNR: JDNDMSR+ exceeds TENet in every listed setting; values are shown in Table 3 | Helped |
| Train with MAE loss instead of MSE loss (ablation) | JDNDMSR+ · McMaster and Kodak (averaged), σ=10, ×2 | cPSNR: 28.48 -> 28.73 | Helped |
| Replace RCAB with RAB (ablation) | JDNDMSR+ feature extraction module · McMaster | cPSNR: 28.97 -> 28.55 | Hurt |
| Replace RCAB with RRDB (ablation) | JDNDMSR+ feature extraction module · McMaster | cPSNR: 28.97 -> 28.59 | Hurt |
| Train with SSIM loss instead of MSE loss (ablation) | JDNDMSR+ · McMaster and Kodak (averaged), σ=10, ×2 | cPSNR: 28.48 -> 26.64 | Hurt |
| Train with MS-SSIM loss instead of MSE loss (ablation) | JDNDMSR+ · McMaster and Kodak (averaged), σ=10, ×2 | cPSNR: 28.48 -> 26.64 | Hurt |
| Train with Mix1 loss instead of MSE loss (ablation) | JDNDMSR+ · McMaster and Kodak (averaged), σ=10, ×2 | cPSNR: 28.48 -> 27.36 | Hurt |
Closest papers by meaning
- Image Restoration Using Very Deep Convolutional Encoder-Decoder Networks with Symmetric Skip Connections (NeurIPS 2016 · similarity 0.90)
- Joint Demosaicing and Denoising With Self Guidance (CVPR 2020 · similarity 0.89)
- Universal Denoising Networks : A Novel CNN Architecture for Image Denoising (CVPR 2018 · similarity 0.89)
- ResDiff: Combining CNN and Diffusion Model for Image Super-resolution (AAAI 2024 · similarity 0.88)
- A Restoration Network as an Implicit Prior (ICLR 2024 · similarity 0.87)
- One-Step Effective Diffusion Network for Real-World Image Super-Resolution (NeurIPS 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.