Research map / Image generation / Image restoration
Online-Updated High-Order Collaborative Networks for Single Image Deraining
Read the paperCheck 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 |
|---|---|---|---|
| Add bidirectional scale-content similarity mining instead of omitting BiSCSM (ablation) | High-order collaborative network · Rain200H | PSNR: 29.583 -> 29.985 | Helped |
| Use bidirectional mining rather than element-wise summation or 1 × 1 convolution for scale fusion (ablation) | BiSCSM · Rain200H | PSNR: 29.625 -> 29.985; 29.541 -> 29.985 | Helped |
| Use both Down-to-Up and Up-to-Down mining rather than either direction alone (ablation) | BiSCSM · Rain200H | PSNR: 29.393 -> 29.985; 29.828 -> 29.985 | Helped |
| Add positional embedding (ablation) | BiSCSM · Rain200H | PSNR: 29.953 -> 29.985 | Helped |
| Add both 1/2- and 1/4-scale Laplacian constraints instead of no constraints (ablation) | High-order collaborative network · Rain200H | PSNR: 29.824 -> 29.985 | Helped |
| Use multi-scale Laplacian constraints rather than a full-scale constraint (ablation) | Multi-scale compact constraints · Rain200H | PSNR: 29.970 -> 29.985 | Helped |
| Add collaborative top and middle sub-networks to the bottom sub-network (ablation) | High-order collaborative network · Rain200H | PSNR: 29.563 -> 29.985 | Helped |
| Train the sub-networks collaboratively rather than cascade them (ablation) | High-order collaborative network · Rain200H | PSNR: 29.642 -> 29.985 | Helped |
| Apply online-update learning with updated pseudo ground truth rather than direct real-world fine-tuning (ablation) | Proposed deraining network · Real-world rainy images | Visual deraining quality | Helped |
| Apply online-update learning to a pretrained synthetic-data deraining model (ablation) | DCSFN · Real-world rainy images | Visual deraining quality | Helped |
Closest papers by meaning
- Multi-Scale Progressive Fusion Network for Single Image Deraining (CVPR 2020 · similarity 0.90)
- Multi-Decoding Deraining Network and Quasi-Sparsity Based Training (CVPR 2021 · similarity 0.89)
- Bidirectional Multi-Scale Implicit Neural Representations for Image Deraining (CVPR 2024 · similarity 0.87)
- Rethinking Multi-Scale Representations in Deep Deraining Transformer (AAAI 2024 · similarity 0.86)
- Channel Consistency Prior and Self-Reconstruction Strategy Based Unsupervised Image Deraining (CVPR 2025 · similarity 0.86)
- Harnessing Joint Rain-/Detail-aware Representations to Eliminate Intricate Rains (ICLR 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.