Research map / Image generation / Super-resolution
Test-Time Training for Hyperspectral Image Super-Resolution
Ke Li, Luc Van Gool, Dengxin Dai
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 |
|---|---|---|---|
| process each spectral band separately with a shared single-channel network | LIIF · CAVE x2 | MPSNR: 38.7380 -> 39.8385 | Improved |
| process each spectral band separately with a shared single-channel network | LIIF · Spectral World x2 | MPSNR: 38.9754 -> 40.2182 | Improved |
| add SARM channel-interaction module (ablation) | LIIF single-channel · CAVE x2 | MPSNR: 39.8385 -> 39.4385 | Hurt |
| add Restormer channel-interaction module (ablation) | LIIF single-channel · CAVE x2 | MPSNR: 39.8385 -> 38.8234 | Hurt |
| add self-training with pseudo samples (ablation) | single-channel LIIF · CAVE x2 | MPSNR: 39.8385 -> 40.8385 | Improved |
| add weight-averaged teacher-student learning (ablation) | LIIF + pseudo samples · Harvard x2 | MPSNR: 40.8256 -> 41.8720 | Improved |
| add Spectral Mixup (ablation) | LIIF + pseudo + teacher · CAVE x2 | MPSNR: 41.2380 -> 41.8654 | Improved |
| add Spectral Mixup (ablation) | LIIF + pseudo + teacher · CAVE x4 | MPSNR: 38.9541 -> 39.2376 | Improved |
| use full test-time training instead of source model | LIIF · CAVE x4 | MPSNR: 37.7645 -> 39.2376 | Improved |
| use full test-time training instead of source model | SwinIR · CAVE x4 | PSNR: 42.0143 -> 43.3564 | Improved |
| use full test-time training instead of source model | HSIwRGB · CAVE x4 | PSNR: 43.3242 -> 44.2489 | Improved |
| use standard mixup instead of Spectral Mixup (ablation) | full method · CAVE x4 | MPSNR: 39.2376 -> 37.2530 | Hurt |
Closest papers by meaning
- EigenSR: Eigenimage-Bridged Pre-Trained RGB Learners for Single Hyperspectral Image Super-Resolution (AAAI 2025 · similarity 0.92)
- Hyperspectral Image Super-Resolution With Optimized RGB Guidance (CVPR 2019 · similarity 0.92)
- Unsupervised Test-Time Adaptation Learning for Effective Hyperspectral Image Super-Resolution With Unknown Degeneration (TPAMI 2024 · similarity 0.91)
- Learning Multi-Modal Cross-Scale Deformable Transformer Network for Unregistered Hyperspectral Image Super-resolution (AAAI 2024 · similarity 0.87)
- Enhancing Hyperspectral Images via Diffusion Model and Group-Autoencoder Super-resolution Network (AAAI 2024 · similarity 0.87)
- 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.
