Research map / Image generation / Image restoration
Hyperspectral Image Reconstruction Using a Deep Spatial-Spectral Prior
Check 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 |
|---|---|---|---|
| Use the learned spatial-spectral prior and optimization-inspired network instead of the strongest listed baseline | CASSI reconstruction · ICVL | PSNR: 31.73 -> 34.13 | Helped |
| Use the learned spatial-spectral prior and optimization-inspired network instead of the strongest listed baseline | CASSI reconstruction · Harvard | PSNR: 29.68 -> 32.84 | Helped |
| Use the learned spatial-spectral prior and optimization-inspired network instead of the strongest listed baseline | CASSI reconstruction · KAIST, without noise | PSNR: 29.39 -> 30.03 | Helped |
| Use the learned spatial-spectral prior and optimization-inspired network instead of the strongest listed baseline | CASSI reconstruction · KAIST, noise variance 0.05 | PSNR: 29.28 -> 29.86 | Helped |
| Use different coded apertures in training and testing rather than identical apertures | ProposedD CASSI reconstruction · ICVL | PSNR: 34.13 -> 33.43 | Hurt |
| Use different coded apertures in training and testing rather than identical apertures | ProposedD CASSI reconstruction · Harvard | PSNR: 32.84 -> 32.44 | Hurt |
| Add the spatial network part to the prior network (ablation) | Hyperspectral image prior network · CASSI reconstruction | reconstruction performance | Helped |
| Add the spectral network part to the prior network (ablation) | Hyperspectral image prior network · CASSI reconstruction | reconstruction performance | Helped |
| Increase the unrolled network stage count beyond 9 (ablation) | Optimization-inspired reconstruction network · CASSI reconstruction | PSNR | No clear effect |
| Increase the unrolled network stage count beyond 9 (ablation) | Optimization-inspired reconstruction network · CASSI reconstruction | SAM | No clear effect |
| Add imaging noise with variance 0.05 | ProposedI CASSI reconstruction · KAIST | PSNR: 30.03 -> 29.86 | Hurt |
| Add imaging noise with variance 0.05 | ProposedI CASSI reconstruction · KAIST | SAM: 0.114 -> 0.136 | Hurt |
Closest papers by meaning
- Learning Tensor Low-Rank Prior for Hyperspectral Image Reconstruction (CVPR 2021 · similarity 0.91)
- SPECAT: SPatial-spEctral Cumulative-Attention Transformer for High-Resolution Hyperspectral Image Reconstruction (CVPR 2024 · similarity 0.88)
- LESSViT: Robust Hyperspectral Representation Learning under Spectral Configuration Shift (NeurIPS 2026 · similarity 0.87)
- FRN: Fractal-Based Recursive Spectral Reconstruction Network (NeurIPS 2025 · similarity 0.87)
- Hyperspectral Image Reconstruction via Combinatorial Embedding of Cross-Channel Spatio-Spectral Clues (AAAI 2024 · similarity 0.87)
- HIR-Diff: Unsupervised Hyperspectral Image Restoration Via Improved Diffusion Models (CVPR 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.