Research map / Image generation / Image restoration
RS-vHeat: Heat Conduction Guided Efficient Remote Sensing Foundation Model
Huiyang Hu, Peijin Wang, Hanbo Bi, Boyuan Tong, Zhaozhi Wang, Wenhui Diao, Hao Chang, Yingchao Feng, Ziqi Zhang, Yaowei Wang, Qixiang Ye, Kun Fu and 1 more
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 frequency domain reconstruction loss to spatial reconstruction (ablation) | RS-vHeat pretraining · DIOR | mAP50: 78.2 -> 79.5 | Improved |
| add contrastive loss on top of SDR and FDR (ablation) | RS-vHeat pretraining · DIOR | mAP50: 79.5 -> 82.3 | Improved |
| add frequency domain reconstruction loss (ablation) | RS-vHeat pretraining · WHU-OPT-SAR | OA: 80.5 -> 82.3 | Improved |
| add contrastive loss (ablation) | RS-vHeat pretraining · WHU-OPT-SAR | OA: 82.3 -> 83.9 | Improved |
| add FDR (ablation) | RS-vHeat pretraining · SAR-AIRcraft-1.0 | mAP50: 84.6 -> 86.4 | Improved |
| add contrastive loss (ablation) | RS-vHeat pretraining · SAR-AIRcraft-1.0 | mAP50: 86.4 -> 87.1 | Improved |
| add FDR (ablation) | RS-vHeat pretraining · iSAID | mIoU: 66.1 -> 67.2 | Improved |
| add contrastive loss (ablation) | RS-vHeat pretraining · iSAID | mIoU: 67.2 -> 68.7 | Improved |
| add FDR (ablation) | RS-vHeat pretraining · AIR-PolSAR-Seg | mIoU: 54.7 -> 55.1 | Improved |
| add contrastive loss (ablation) | RS-vHeat pretraining · AIR-PolSAR-Seg | mIoU: 55.1 -> 57.5 | Improved |
| replace attention-based Swin-B RSFMs with heat conduction operator backbone | RS-vHeat · 1024x1024 images | throughput and memory: 2.7 times throughput, 84% less memory | Improved |
| use RS-vHeat instead of SkySense (Swin-H) | RS-vHeat · iSAID | mIoU: 70.91 -> 68.72 | Hurt |
Closest papers by meaning
- Parameter Efficient Self-Supervised Geospatial Domain Adaptation (CVPR 2024 · similarity 0.85)
- SkySense: A Multi-Modal Remote Sensing Foundation Model Towards Universal Interpretation for Earth Observation Imagery (CVPR 2024 · similarity 0.84)
- Building Vision Models upon Heat Conduction (CVPR 2025 · similarity 0.84)
- Pan-LUT: Efficient Pan-sharpening via Learnable Look-Up Tables (NeurIPS 2025 · similarity 0.84)
- MMM-RS: A Multi-modal, Multi-GSD, Multi-scene Remote Sensing Dataset and Benchmark for Text-to-Image Generation (NeurIPS 2024 · similarity 0.83)
- Meta-Learning Hyperparameters for Parameter Efficient Fine-Tuning (CVPR 2025 · similarity 0.83)
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.
