Research map / Image generation / Image restoration
Rotation Equivariant Proximal Operator for Deep Unfolding Methods in Image Restoration
Jiahong Fu, Qi Xie, Deyu Meng, Zongben Xu
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 |
|---|---|---|---|
| replace standard convolution with F-Conv equivariant convolution in proximal network (ablation) | Prox_CNN · Urban100 denoising | PSNR: 30.57 -> 31.05 | Improved |
| use PDO-eConv equivariant convolution instead of standard CNN (ablation) | Prox_CNN · Urban100 denoising | PSNR: 30.57 -> 29.55 | Hurt |
| use G-CNN or E2-CNN equivariant convolution instead of standard CNN (ablation) | Prox_CNN · Urban100 denoising | PSNR: 30.57 -> 30.75 | No clear effect |
| increase equivariant number t from 1 to 24 (ablation) | Rotation equivariant CNN · DIV2K | relative equivariant error: 0.852 -> 0.068 | Improved |
| add rotation equivariant proximal network (p8+) to KXNet | KXNet · Urban100 SR x4 (std 0) | PSNR: 25.41 -> 25.69 | Improved |
| add rotation equivariant proximal network (p8+) to KXNet on out-of-distribution noise | KXNet · Urban100 SR x4 (std 50) | PSNR: 21.83 -> 22.06 | Improved |
| use p8 (fewer channels, t=8) instead of p4 for KXNet (ablation) | KXNet-p4 · Urban100 SR x4 (std 0) | PSNR: 25.55 -> 25.30 | Hurt |
| add rotation equivariant proximal network (p8+) to ACDNet | ACDNet · DeepLesion MAR | PSNR: 41.44 -> 42.43 | Improved |
| add rotation equivariant proximal network (p8+) to DICDNet | DICDNet · DeepLesion MAR | PSNR: 42.24 -> 43.52 | Improved |
| add rotation equivariant proximal network (p8+) to OSCNet | OSCNet · DeepLesion MAR | PSNR: 42.21 -> 42.90 | Improved |
| use p8 instead of p4 for DICDNet (ablation) | DICDNet-p4 · DeepLesion MAR | PSNR: 42.67 -> 42.48 | No clear effect |
| add rotation equivariant proximal network (p8+) to RCDNet | RCDNet · Rain100H | PSNR: 31.28 -> 31.81 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- KXNet — KXNet: A Model-Driven Deep Neural Network for Blind Super-Resolution (ECCV 2022) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- A Regularization-Guided Equivariant Approach for Image Restoration (CVPR 2025 · similarity 0.89)
- What's in a Prior? Learned Proximal Networks for Inverse Problems (ICLR 2024 · similarity 0.87)
- Rotation-Equivariant Self-Supervised Method in Image Denoising (CVPR 2025 · similarity 0.86)
- On the Fourier analysis in the SO(3) space : the EquiLoPO Network (ICLR 2025 · similarity 0.85)
- Proximal Algorithm Unrolling: Flexible and Efficient Reconstruction Networks for Single-Pixel Imaging (CVPR 2025 · similarity 0.85)
- Probing Equivariance and Symmetry Breaking in Convolutional Networks (NeurIPS 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.
