Research map / Image generation / Image restoration
Deep Gaussian Conditional Random Field Network: A Model-Based Deep Network for Discriminative Denoising
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 the deep GCRF network rather than WNNM | image denoising · 300 test images, σ=10–60 | PSNR: 0.29 - 1.24 dB | Helped |
| use the deep GCRF network rather than WNNM | image denoising · 300 test images, σ=10 | PSNR: 33.57; 33.56 | No clear effect |
| use the deep GCRF network rather than BM3D | image denoising · 300 test images, σ=10–60 | PSNR: 0.18 - 0.83 dB | Helped |
| use the deep GCRF network rather than TRD | image denoising · 68-image unquantized setting, σ=15 and σ=25 | PSNR: 0.07 dB; 0.11 dB | Mixed |
| use the deep GCRF network rather than MLP | image denoising · 300 test images, σ=10–50 reported comparisons | PSNR | Mixed |
| use one noise-aware model for a range of noise levels instead of separate noise-specific MLPs | deep GCRF denoising · noise levels including σ=10, 15, 20, 45, 55, and 60 | PSNR | Helped |
| use 8×8 rather than 5×5 patches (ablation) | deep GCRF · 300 test images, σ=10–60 | PSNR: 33.53; 33.56; 31.29; 31.35 | Helped |
| incorporate input noise variance in the quadratic selection function | deep GCRF parameter generation network · multiple noise levels | PSNR | Helped |
| add parameter-generation networks after HQS iterations, operating on partially restored images | deep GCRF · high-noise image denoising | PSNR | Helped |
| train the base potential matrices discriminatively rather than fitting them generatively with a GMM | deep GCRF versus EPLL · image denoising | PSNR: 0.24 - 1.18 dB | Helped |
Closest papers by meaning
- Universal Denoising Networks : A Novel CNN Architecture for Image Denoising (CVPR 2018 · similarity 0.91)
- Robust And Interpretable Blind Image Denoising Via Bias-Free Convolutional Neural Networks (ICLR 2020 · similarity 0.90)
- An Unsupervised Deep Learning Approach for Real-World Image Denoising (ICLR 2021 · similarity 0.89)
- Gaussian Conditional Random Field Network for Semantic Segmentation (CVPR 2016 · similarity 0.89)
- Variational Denoising Network: Toward Blind Noise Modeling and Removal (NeurIPS 2019 · similarity 0.89)
- Zero-Shot Noise2Mean: Gap Minimization for Efficient Denoising from a Single Noisy Image (AAAI 2025 · similarity 0.87)
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.