Research map / Image generation / Image restoration
Ingredient-Oriented Multi-Degradation Learning for Image Restoration
Jinghao Zhang, Jie Huang, Mingde Yao, Zizheng Yang, Hu Yu, Man Zhou, Feng Zhao
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 |
|---|---|---|---|
| Use ingredient-oriented degradation reformulation with task-specific prior collection followed by learnable PCA integration | Restormer-based IDR · Rain100L, SOTS, BSD68, GoPro, and LOL | PSNR: Rain100L: 35.63 vs 38.16; SOTS: 25.24 vs 25.23; BSD68: 31.60 vs 31.49; GoPro: 27.87 vs 26.87; LOL: 21.34 vs 21.87 | Mixed |
| Use IDR without fine-tuning on an unknown downstream task | IDR · TOLED and POLED UDC restoration | PSNR: TOLED: 27.91 vs 26.89; POLED: 16.71 vs 13.92 | Improved |
| Include task-oriented prior learning before ingredient-oriented integration (ablation) | IDR · Combined five-task image restoration dataset | PSNR | Improved |
| Use learnable PCA instead of random integration (ablation) | IDR · Combined five-task image restoration dataset | PSNR | Improved |
| Use dynamic routing (ablation) | IDR · Combined five-task image restoration dataset | PSNR | Improved |
| Use supervised degradation attention (ablation) | IDR · Combined five-task image restoration dataset | PSNR | Improved |
| Train with five restoration tasks rather than the R+H+N subset (ablation) | IDR · LOL | PSNR: 7.82 -> 21.34 | Improved |
| Train with five restoration tasks rather than the R+H+N+B subset (ablation) | IDR · GoPro | PSNR: 27.08 -> 27.87 | Improved |
| Train with five restoration tasks rather than the R+H+N+L subset (ablation) | IDR · GoPro | PSNR: 25.50 -> 27.87 | 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.
- Restormer — Restormer: Efficient Transformer for High-Resolution Image Restoration (CVPR 2022) · better on 6 of 7 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
- Learning Where and What to Restore for Composite Image Restoration (NeurIPS 2026 · similarity 0.90)
- Continuous Expert Assembly: Instance-Conditioned Low-Rank Residuals for All-in-One Image Restoration (NeurIPS 2026 · similarity 0.89)
- Debiased All-in-one Image Restoration with Task Uncertainty Regularization (AAAI 2025 · similarity 0.88)
- Complexity Experts are Task-Discriminative Learners for Any Image Restoration (CVPR 2025 · similarity 0.88)
- Visual-Instructed Degradation Diffusion for All-in-One Image Restoration (CVPR 2025 · similarity 0.88)
- Learning from History: Task-agnostic Model Contrastive Learning for Image Restoration (AAAI 2024 · 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.
