Research map / Image generation / Super-resolution
Learning Mutual Modulation for Self-Supervised Cross-Modal Super-Resolution
Xiaoyu Dong, Naoto Yokoya, Longguang Wang, Tatsumi Uezato
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 guide-to-source modulation (ablation) | MMSR · Middlebury 2014 x8 depth SR | RMSE: 4.30 -> 4.08 | Improved |
| add source-to-guide modulation (ablation) | MMSR · Middlebury 2014 x8 depth SR | RMSE: 4.30 -> 3.72 | Improved |
| add both modulations (mutual) (ablation) | MMSR · Middlebury 2014 x8 depth SR | RMSE: 4.30 -> 3.67 | Improved |
| add cross-domain filtering to adaptive filters (ablation) | MMSR · Middlebury 2014 x8 depth SR | RMSE: 3.84 -> 3.67 | Improved |
| add adaptive filtering to cross-domain filters (ablation) | MMSR · Middlebury 2014 x8 depth SR | RMSE: 4.88 -> 3.67 | Improved |
| add cross-domain filtering alone (ablation) | MMSR · Middlebury 2014 x8 depth SR | RMSE: 4.87 -> 4.88 | No clear effect |
| enlarge neighborhood size from 3x3 to 11x11 in both modulations (ablation) | MMSR · Middlebury 2014 x8 depth SR | RMSE: 4.23 -> 3.67 | Improved |
| enlarge neighborhood size from 7x7 to 11x11 in both modulations (ablation) | MMSR · Middlebury 2014 x8 depth SR | RMSE: 3.91 -> 3.67 | Improved |
| fuse with 1x1 conv instead of naive summation (ablation) | MMSR · Middlebury 2003 x4 depth SR | RMSE: 1.88 -> 1.78 | Improved |
| fuse with 1x1 conv instead of attention plus summation (ablation) | MMSR · Middlebury 2003 x4 depth SR | RMSE: 1.92 -> 1.78 | Improved |
| fuse with 1x1 conv instead of attention plus 1x1 conv (ablation) | MMSR · Middlebury 2003 x4 depth SR | RMSE: 1.84 -> 1.78 | Improved |
Closest papers by meaning
- Robust Self-Supervised Cross-Modal Super-Resolution against Real-World Misaligned Observations (ECCV 2026 · similarity 0.91)
- Single Pair Cross-Modality Super Resolution (CVPR 2021 · similarity 0.91)
- Low-Res Leads the Way: Improving Generalization for Super-Resolution by Self-Supervised Learning (CVPR 2024 · similarity 0.86)
- Event-Guided Super-Resolving Blurry Image via Asymmetric Integral Driven Consistency (AAAI 2026 · similarity 0.86)
- Recursive Generalization Transformer for Image Super-Resolution (ICLR 2024 · similarity 0.84)
- Pixel-level and Semantic-level Adjustable Super-resolution: A Dual-LoRA Approach (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.
