Research map / Architectures and efficiency / Attention and transformers
Spatial-Spectral Transformer for Hyperspectral Image Denoising
Miaoyu Li, Ying Fu, Yulun 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 |
|---|---|---|---|
| use spatial non-local and global spectral attention together (ablation) | SST · ICVL, Gaussian noise σ=50 | PSNR: 40.69 -> 41.09 | Improved |
| add NLSA (ablation) | SST · ICVL, Gaussian noise σ=50 | PSNR: 34.67 -> 41.09 | Improved |
| add GSA (ablation) | SST · ICVL, Gaussian noise σ=50 | PSNR: 40.44 -> 41.09 | Improved |
| replace GSA with NLSA (ablation) | SST · ICVL, Gaussian noise σ=50 | PSNR: 41.09 -> 40.56 | Hurt |
| replace NLSA with GSA (ablation) | SST · ICVL, Gaussian noise σ=50 | PSNR: 41.09 -> 39.82 | Hurt |
| reverse the attention order to GSA then NLSA (ablation) | SST · ICVL, Gaussian noise σ=50 | PSNR: 41.09 -> 40.69 | Hurt |
| increase the NLSA window size from 2 to 8 (ablation) | SST · ICVL, Gaussian noise σ=50 | PSNR: 42.13 -> 42.57 | Improved |
| increase the NLSA window size from 8 to 16 (ablation) | SST · ICVL, Gaussian noise σ=50 | PSNR: 42.57 -> 42.59; 20.70 -> 30.24 GFLOPs | Mixed |
| use SST instead of the strongest listed baseline | HSI denoising · ICVL, Gaussian noise σ=10–70 | PSNR: 41.67 -> 42.57 | Improved |
| use SST instead of QRNN3D | HSI denoising · ICVL, mixture noise | PSNR: 39.08 -> 38.78 | Hurt |
Closest papers by meaning
- Hybrid Spectral Denoising Transformer with Guided Attention (ICCV 2023 · similarity 0.94)
- Improving Spectral Snapshot Reconstruction with Spectral-Spatial Rectification (CVPR 2024 · similarity 0.89)
- SPECAT: SPatial-spEctral Cumulative-Attention Transformer for High-Resolution Hyperspectral Image Reconstruction (CVPR 2024 · similarity 0.89)
- VolFormer: Explore More Comprehensive Cube Interaction for Hyperspectral Image Restoration and Beyond (CVPR 2025 · similarity 0.89)
- Dual-Window Multiscale Transformer for Hyperspectral Snapshot Compressive Imaging (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.
