Research map / Image generation / Super-resolution
Unsupervised Hyperspectral Image Super-Resolution via Self-Supervised Modality Decoupling
Songcheng Du, Yang Zou, Zixu Wang, X. Rong Li, Ying Li, Changjing Shang, Qiang Shen
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 subspace clustering loss L_SC (ablation) | MossFuse · CAVE | PSNR: 37.83 -> 42.15 | Improved |
| add repulsion term in L_SC (ablation) | MossFuse · CAVE | PSNR: 38.74 -> 42.15 | Improved |
| add aggregation term in L_SC (ablation) | MossFuse · CAVE | PSNR: 39.86 -> 42.15 | Improved |
| use LK-CNN block instead of residual block (ablation) | MossFuse · CAVE | PSNR: 39.36 -> 42.15 | Improved |
| use LK-CNN instead of dilated conv (ablation) | MossFuse · CAVE | PSNR: 40.06 -> 42.15 | Improved |
| use LK-CNN instead of SWT block (ablation) | MossFuse · CAVE | PSNR: 38.76 -> 42.15 | Improved |
| use spatial/spectral-aware aggregation instead of concat+conv (ablation) | MossFuse · CAVE | PSNR: 38.92 -> 42.15 | Improved |
| use aggregation blocks instead of cross-attention (ablation) | MossFuse · CAVE | PSNR: 40.29 -> 42.15 | Improved |
| add self-supervised constraint loss L_SCT (ablation) | MossFuse · CAVE | PSNR: 39.72 -> 42.15 | Improved |
| add modality decoupling (ablation) | MossFuse · CAVE | PSNR: 38.63 -> 42.15 | Improved |
| add modality decoupling (ablation) | MossFuse · Harvard | PSNR: 39.32 -> 44.62 | Improved |
| use proposed degradation estimation instead of conv DE (ablation) | MossFuse · CAVE | PSNR: 39.34 -> 42.15 | Improved |
Closest papers by meaning
- Unsupervised Sparse Dirichlet-Net for Hyperspectral Image Super-Resolution (CVPR 2018 · similarity 0.90)
- Hyperspectral and Multispectral Image Fusion with Arbitrary Resolution Through Self-Supervised Representations (IJCV 2025 · similarity 0.90)
- Enhancing Unregistered Hyperspectral Image Super-Resolution via Unmixing-based Abundance Fusion Learning (CVPR 2026 · similarity 0.90)
- Learning Multi-Modal Cross-Scale Deformable Transformer Network for Unregistered Hyperspectral Image Super-resolution (AAAI 2024 · similarity 0.88)
- EigenSR: Eigenimage-Bridged Pre-Trained RGB Learners for Single Hyperspectral Image Super-Resolution (AAAI 2025 · similarity 0.88)
- S2CycleDiff: Spatial-Spectral-Bilateral Cycle-Diffusion Framework for Hyperspectral Image Super-resolution (AAAI 2024 · 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.
