Research map / Image generation / Image restoration
Spectral Enhanced Rectangle Transformer for Hyperspectral Image Denoising
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 spectral enhancement to rectangle attention (ablation) | SERT · ICVL, random Gaussian noise | PSNR: 42.06 -> 42.54 | Helped |
| Add spectral shuffle to rectangle attention and spectral enhancement (ablation) | SERT · ICVL, random Gaussian noise | PSNR: 42.54 -> 42.60 | Helped |
| Add memory unit to rectangle attention, spectral enhancement, and spectral shuffle (ablation) | SERT · ICVL, random Gaussian noise | PSNR: 42.60 -> 42.82 | Helped |
| Use global SE over the whole HSI instead of non-local SE over neighboring rectangles (ablation) | SERT · ICVL, random Gaussian noise | PSNR: 42.82 -> 42.04 | Hurt |
| Use local SE within each rectangle instead of non-local SE over neighboring rectangles (ablation) | SERT · ICVL, random Gaussian noise | PSNR: 42.82 -> 42.60 | Hurt |
| Use SERT instead of QRNN3D | HSI denoising · ICVL, Gaussian+Mixture | PSNR: 39.08 -> 40.00 | Helped |
| Use SERT instead of MAC-Net | HSI denoising · Realistic dataset, 15 real noisy HSIs | PSNR: 29.20 -> 29.68 | Helped |
| Use longer rectangle lengths (ablation) | SERT · ICVL, random Gaussian noise | PSNR | Mixed |
Closest papers by meaning
- Improving Spectral Snapshot Reconstruction with Spectral-Spatial Rectification (CVPR 2024 · similarity 0.88)
- HIR-Diff: Unsupervised Hyperspectral Image Restoration Via Improved Diffusion Models (CVPR 2024 · similarity 0.88)
- Unmixing Diffusion for Self-Supervised Hyperspectral Image Denoising (CVPR 2024 · similarity 0.87)
- SPECAT: SPatial-spEctral Cumulative-Attention Transformer for High-Resolution Hyperspectral Image Reconstruction (CVPR 2024 · similarity 0.87)
- LESSViT: Robust Hyperspectral Representation Learning under Spectral Configuration Shift (NeurIPS 2026 · similarity 0.87)
- VolFormer: Explore More Comprehensive Cube Interaction for Hyperspectral Image Restoration and Beyond (CVPR 2025 · 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.