Research map / Image generation / Image restoration
MP-HSIR: A Multi-Prompt Framework for Universal Hyperspectral Image Restoration
Zhehui Wu, Yong Chen, Naoto Yokoya, Wei He
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 textual prompt (ablation) | spatial-SA baseline · Chikusei inpainting (mask 0.9) | PSNR: 39.24 -> 39.62 | Improved |
| add learnable visual prompt (ablation) | spatial-SA baseline · Chikusei inpainting (mask 0.9) | PSNR: 39.24 -> 39.57 | Improved |
| add both textual and visual prompts (ablation) | spatial-SA baseline · Chikusei inpainting (mask 0.9) | PSNR: 39.24 -> 39.90 | Improved |
| add global spectral self-attention (ablation) | spatial SA with text-visual prompts · Chikusei inpainting (mask 0.9) | PSNR: 39.90 -> 40.63 | Improved |
| add local spectral self-attention (ablation) | spatial SA with text-visual prompts · Chikusei inpainting (mask 0.9) | PSNR: 39.90 -> 40.46 | Improved |
| add spectral prompt to local spectral attention (ablation) | local spectral SA with text-visual prompts · Chikusei inpainting (mask 0.9) | PSNR: 40.46 -> 41.05 | Improved |
| combine global and local spectral branches with spectral prompt (ablation) | MP-HSIR · Chikusei inpainting (mask 0.9) | PSNR: 41.05 -> 41.98 | Improved |
| use MP-HSIR instead of best prior method | inpainting · Inpainting ICVL | PSNR: 46.38 -> 51.53 | Improved |
| use MP-HSIR instead of best prior method | band completion · Band completion Berlin | PSNR: 43.36 -> 49.50 | Improved |
| use MP-HSIR instead of best task-specific method | Gaussian denoising · Gaussian denoising ICVL | PSNR: 41.92 -> 41.62 | Hurt |
| use MP-HSIR instead of best task-specific method | complex denoising · Complex denoising ICVL | PSNR: 43.42 -> 42.29 | Hurt |
| fine-tune MP-HSIR with 5% data on unseen task vs best task-specific | motion deblurring · Motion deblurring ICVL | PSNR: 39.28 -> 40.06 | Improved |
Later papers that compared with this method
1 later accepted paper reports MP-HSIR as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Degradation-Aware Metric Prompting for Hyperspectral Image Restoration (ICML 2026 · better on 4 of 4 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- PromptIR — PromptIR: Prompting for All-in-One Image Restoration (NeurIPS 2023) · better on 2 of 2 shared results
- InstructIR — InstructIR: High-Quality Image Restoration Following Human Instructions (ECCV 2024) · better on 1 of 1 shared result
- SRFormer — SRFormer: Permuted Self-Attention for Single Image Super-Resolution (ICCV 2023) · better on 1 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
- Degradation-Aware Metric Prompting for Hyperspectral Image Restoration (ICML 2026 · similarity 0.91)
- Spectrum-Aware and Transferable Architecture Search for Hyperspectral Image Restoration (ECCV 2022 · similarity 0.90)
- HIR-Diff: Unsupervised Hyperspectral Image Restoration Via Improved Diffusion Models (CVPR 2024 · similarity 0.90)
- Latent Diffusion Enhanced Rectangle Transformer for Hyperspectral Image Restoration (TPAMI 2024 · similarity 0.90)
- Improving Spectral Snapshot Reconstruction with Spectral-Spatial Rectification (CVPR 2024 · similarity 0.88)
- SPECAT: SPatial-spEctral Cumulative-Attention Transformer for High-Resolution Hyperspectral Image Reconstruction (CVPR 2024 · similarity 0.88)
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.
