Research map / Image generation / Image restoration
Stacking Networks Dynamically for Image Restoration Based on the Plug-and-Play Framework
Haixin Wang, Tianhao Zhang, Muzhi Yu, Jinan Sun, Wei Ye, Chen Wang, Shikun Zhang
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 |
|---|---|---|---|
| wrap pre-trained DMPHN(1-2-4) in the dynamic plug-and-play framework | DMPHN(1-2-4) · GoPro deblurring | PSNR (dB): 30.21 -> 30.32 | Improved |
| wrap pre-trained DMPHN(1-2-4-8) in the framework | DMPHN(1-2-4-8) · GoPro deblurring | PSNR (dB): 30.25 -> 30.40 | Improved |
| wrap Stack(2)-DMPHN in the framework | Stack(2)-DMPHN · GoPro deblurring | PSNR (dB): 30.71 -> 30.92 | Improved |
| wrap Stack(3)-DMPHN in the framework | Stack(3)-DMPHN · GoPro deblurring | PSNR (dB): 31.16 -> 31.32 | Improved |
| wrap Stack(4)-DMPHN in the framework | Stack(4)-DMPHN · GoPro deblurring | PSNR (dB): 31.20 -> 31.44 | Improved |
| wrap Stack(2)-VMPHN in the framework | Stack(2)-VMPHN · GoPro deblurring | PSNR (dB): 31.50 -> 31.72 | Improved |
| wrap VDN denoiser in the framework (Set12, sigma=50) | VDN · Set12 denoising sigma=50 | PSNR (dB): 28.00 -> 28.27 | Improved |
| wrap VDN denoiser in the framework (BSD68, sigma=50) | VDN · BSD68 denoising sigma=50 | PSNR (dB): 27.18 -> 27.64 | Improved |
| wrap VDN denoiser in the framework (Set12, sigma=15) | VDN · Set12 denoising sigma=15 | SSIM: 0.912 -> 0.912 | No clear effect |
| use shallower Stack(3) with framework instead of deeper Stack(4) alone | Proposed(Stack(3)-DMPHN) · GoPro deblurring | PSNR (dB): 31.20 -> 31.32 | Mixed |
| run dynamic framework for a fifth iteration instead of four (ablation) | Proposed(Stack(4)-DMPHN) · GoPro image example | PSNR (dB): 32.16 -> 31.98 | Hurt |
| increase dynamic iterations from one to four (ablation) | Proposed(Stack(4)-DMPHN) · GoPro image example | PSNR (dB): 31.89 -> 32.16 | Improved |
Closest papers by meaning
- Plug-and-Play Image Restoration With Deep Denoiser Prior (TPAMI 2021 · similarity 0.88)
- Denoising Prior Driven Deep Neural Network for Image Restoration (TPAMI 2018 · similarity 0.87)
- Deep Stacked Hierarchical Multi-Patch Network for Image Deblurring (CVPR 2019 · similarity 0.87)
- A Comparative Study of Image Restoration Networks for General Backbone Network Design (ECCV 2024 · similarity 0.87)
- Neural Residual Diffusion Models for Deep Scalable Vision Generation (NeurIPS 2024 · similarity 0.85)
- Vision-Language Gradient Descent-driven All-in-One Deep Unfolding Networks (CVPR 2025 · similarity 0.85)
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.
