Research map / Image generation / Image restoration
M3SR: Multi-Scale Multi-Perceptual Mamba for Efficient Spectral Reconstruction
Yuze Zhang, Lingjie Li, Qiuzhen Lin, Zhong Ming, Fei Yu, Victor C. M. Leung
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 spatial perceptual branch (Mamba SS2D) to MPF block (ablation) | M3SR · NTIRE2022 | PSNR: 30.4932 -> 31.3995 | Improved |
| add spatial perceptual branch (Mamba SS2D) to MPF block (ablation) | M3SR · NTIRE2022 | SAM: 12.5459 -> 6.6199 | Improved |
| add frequency (DWT) perceptual branch (ablation) | M3SR · NTIRE2022 | PSNR: 30.5882 -> 31.3995 | Improved |
| add frequency (DWT) perceptual branch (ablation) | M3SR · NTIRE2022 | SAM: 6.52 -> 6.6199 | Hurt |
| add spectral perceptual branch (Mamba across bands) (ablation) | M3SR · NTIRE2022 | PSNR: 30.3648 -> 31.3995 | Improved |
| add spectral perceptual branch (Mamba across bands) (ablation) | M3SR · NTIRE2022 | SAM: 6.2789 -> 6.6199 | Hurt |
| use 4 spectral groups instead of 2 (ablation) | M3SR · NTIRE2022 | PSNR: 31.2306 -> 31.3995 | Improved |
| increase spectral groups from 4 to 8 (ablation) | M3SR · NTIRE2022 | PSNR: 31.3995 -> 30.6399 | Hurt |
| increase spectral groups from 4 to 16 (ablation) | M3SR · NTIRE2022 | PSNR: 31.3995 -> 30.7574 | Hurt |
| use M3SR instead of MST++ (Mamba multi-perceptual U-Net vs Transformer) | spectral reconstruction · NTIRE2022 | FLOPs (G): 177.737 -> 100.924 | Improved |
| use M3SR instead of MST++ | spectral reconstruction · CAVE | PSNR: 34.6534 -> 35.6063 | Improved |
| use M3SR instead of GMSR (lighter Mamba baseline) | spectral reconstruction · NTIRE2022 | PSNR: 26.9156 -> 31.3995 | Improved |
Closest papers by meaning
- Sp3ctralMamba: Physics-Driven Joint State Space Model for Hyperspectral Image Reconstruction (AAAI 2025 · similarity 0.88)
- Detail Matters: Mamba-Inspired Joint Unfolding Network for Snapshot Spectral Compressive Imaging (AAAI 2025 · similarity 0.87)
- EigenSR: Eigenimage-Bridged Pre-Trained RGB Learners for Single Hyperspectral Image Super-Resolution (AAAI 2025 · similarity 0.85)
- Hyperspectral Image Reconstruction via Combinatorial Embedding of Cross-Channel Spatio-Spectral Clues (AAAI 2024 · similarity 0.85)
- SPECAT: SPatial-spEctral Cumulative-Attention Transformer for High-Resolution Hyperspectral Image Reconstruction (CVPR 2024 · similarity 0.85)
- A Selective Re-learning Mechanism for Hyperspectral Fusion Imaging (CVPR 2025 · similarity 0.84)
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.
