Research map / Image generation / Image restoration
Deep Boosting for Image Denoising
Chang Chen, Zhiwei Xiong, Xinmei Tian, Feng Wu
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 |
|---|---|---|---|
| remove the subtraction of the previous estimate (no identical subtraction as in SOS) (ablation) | 3-stage DBF · grey denoising | PSNR: +0.12 dB | Improved |
| increase boosting stages from 1 to 5 (ablation) | DBF · Set12 sigma=50 | PSNR: 0.30 dB gain | Improved |
| greedy training instead of joint training (ablation) | 3-stage DBF · grey denoising | PSNR | No clear effect |
| use deeper thinner plain network (PN2) instead of PN at equal parameters (ablation) | PN · Set12 sigma=50 | PSNR | Improved |
| put PN2 into a 2-stage boosting framework (ablation) | PN2 · Set12 sigma=50 | PSNR | No clear effect |
| add dense connection (ablation) | PN2 · Set12 sigma=50 | convergence and PSNR | Improved |
| add dilated convolution (ablation) | DN · Set12 sigma=50 | PSNR | Improved |
| add path-widening fusion (ablation) | DDN · Set12 sigma=50 | PSNR | Improved |
| double the number of filters in each layer (W) (ablation) | DDFN · Set12 sigma=50 | PSNR | Improved |
| use DDFN-x5W boosting instead of BM3D with SOS boosting | BM3D-SOS · Set5 | PSNR: +0.77 dB (sigma=25), +0.92 dB (sigma=50) | Improved |
| apply SOS boosting to classic denoisers | BM3D · Set5 | PSNR: 31.58 -> 31.61 | Improved |
| use DDFN-x5W instead of DnCNN | DnCNN · Set12 sigma=50 | PSNR: +0.28 dB | Improved |
Closest papers by meaning
- Real-World Image Denoising with Deep Boosting (TPAMI 2019 · similarity 0.94)
- Universal Denoising Networks : A Novel CNN Architecture for Image Denoising (CVPR 2018 · similarity 0.87)
- Deep Gaussian Conditional Random Field Network: A Model-Based Deep Network for Discriminative Denoising (CVPR 2016 · similarity 0.87)
- Unpaired Learning of Deep Image Denoising (ECCV 2020 · similarity 0.87)
- Robust And Interpretable Blind Image Denoising Via Bias-Free Convolutional Neural Networks (ICLR 2020 · similarity 0.86)
- Non-Local Color Image Denoising With Convolutional Neural Networks (CVPR 2017 · 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.
