Research map / Image generation / Image restoration
ClearAIR: A Human-Visual-Perception-Inspired All-in-One Image Restoration
Xu Zhang, Huan Zhang, Guoli Wang, Qian Zhang, Lefei 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 |
|---|---|---|---|
| add Internal Clue Reuse Mechanism (ICRM) (ablation) | ClearAIR · Rain100L (100K iterations) | PSNR (dB): 38.03 -> 38.21 | Improved |
| add Task Identifier (ablation) | ClearAIR · Rain100L (100K iterations) | PSNR (dB): 38.05 -> 38.21 | Improved |
| add semantic guidance unit (SGU) (ablation) | ClearAIR · Rain100L (100K iterations) | PSNR (dB): 37.93 -> 38.21 | Improved |
| add MLLM-based IQA quality guidance (ablation) | ClearAIR · Rain100L (100K iterations) | PSNR (dB): 37.87 -> 38.21 | Improved |
| replace two of three structured priors with learnable parameters (ablation) | ClearAIR · Rain100L (100K iterations) | PSNR (dB): 38.21 -> 37.43 (SGU only) | Hurt |
| use perception order How-Where-What instead of Where-What-How (ablation) | ClearAIR · Rain100L (100K iterations) | PSNR (dB): 37.89 -> 38.21 | Improved |
| use perception order How-Where-What instead of What-How-Where (ablation) | ClearAIR · Rain100L (100K iterations) | PSNR (dB): 38.04 -> 38.21 | Improved |
| ClearAIR vs prior all-in-one methods on three degradations | Restormer backbone · Three Degradations average | PSNR (dB): 32.70 -> 32.74 (vs VLU-Net) | Improved |
| ClearAIR on low-light enhancement vs InstructIR | ClearAIR · LOLv1 (Five Degradations) | PSNR (dB): 23.00 vs 22.83 | Hurt |
| ClearAIR vs OneRestore on composited degradations | ClearAIR · CDD-11 | PSNR (dB): 28.72 -> 29.34 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 3 on most shared results.
- AdaIR — AdaIR: Adaptive All-in-One Image Restoration via Frequency Mining and Modulation (ICLR 2025) · better on 2 of 2 shared results
- OneRestore — OneRestore: A Universal Restoration Framework for Composite Degradation (ECCV 2024) · better on 1 of 1 shared result
- InstructIR — InstructIR: High-Quality Image Restoration Following Human Instructions (ECCV 2024) · better on 0 of 1 shared result
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
- UARE: A Unified Vision-Language Model for Image Quality Assessment, Restoration, and Enhancement (CVPR 2026 · similarity 0.89)
- Continuous Expert Assembly: Instance-Conditioned Low-Rank Residuals for All-in-One Image Restoration (NeurIPS 2026 · similarity 0.88)
- ReFIR: Grounding Large Restoration Models with Retrieval Augmentation (NeurIPS 2024 · similarity 0.88)
- Visual-Instructed Degradation Diffusion for All-in-One Image Restoration (CVPR 2025 · similarity 0.88)
- InstructRestore: Region-Customized Image Restoration with Human Instructions (NeurIPS 2025 · similarity 0.87)
- Learning Where and What to Restore for Composite Image Restoration (NeurIPS 2026 · 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.
