Research map / Label-efficient and robust learning / Recognition and long-tail
Large Scale High-Resolution Land Cover Mapping With Multi-Resolution Data
Caleb Robinson, Le Hou, Kolya Malkin, Rachel Soobitsky, Jacob Czawlytko, Bistra Dilkina, Nebojsa Jojic
Check 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 Color augmentation to SR (ablation) | U-Net · North Chesapeake Test Set | Accuracy: 84.72% -> 85.11% | Improved |
| Add Color augmentation to SR (ablation) | U-Net · Iowa Test Set | Accuracy: 80.91% -> 86.50% | Improved |
| Add LR input augmentation to SR + Color (ablation) | U-Net · North Chesapeake Test Set | Accuracy: 85.11% -> 88.45% | Improved |
| Add LR input augmentation to SR + Color (ablation) | U-Net · Iowa Test Set | Accuracy: 86.50% -> 90.95% | Improved |
| Add Label Overloading to SR + LR + Color (ablation) | U-Net trained in Maryland · North Chesapeake Test Set | Accuracy: 88.45% -> 89.52% | Improved |
| Add Label Overloading to SR + LR + Color (ablation) | U-Net trained in Maryland · Iowa Test Set | Accuracy: 90.95% -> 92.36% | Improved |
| Add Label Overloading to SR + LR + Color (ablation) | U-Net trained on South Chesapeake · Iowa Test Set | Accuracy: 93.07% -> 93.06% | No clear effect |
| Increase model capacity from U-Net to U-Net Large (ablation) | SR + LR + Color + LO · Iowa Test Set | Accuracy: 92.36% -> 92.93% | Improved |
| Use domain-adversarial adaptation (ablation) | U-Net trained in Maryland without data fusion · North Chesapeake Test Set | Accuracy: 61.19% -> 63.33% | Improved |
| Use domain-adversarial adaptation (ablation) | U-Net trained on South Chesapeake without data fusion · Iowa Test Set | Accuracy: 71.00% -> 62.95% | Hurt |
Closest papers by meaning
- Learning without Exact Guidance: Updating Large-scale High-resolution Land Cover Maps from Low-resolution Historical Labels (CVPR 2024 · similarity 0.90)
- GeoLLaVA-8K: Scaling Remote-Sensing Multimodal Large Language Models to 8K Resolution (NeurIPS 2025 · similarity 0.85)
- Rethinking Transformers Pre-training for Multi-Spectral Satellite Imagery (CVPR 2024 · similarity 0.84)
- GeoLink: Empowering Remote Sensing Foundation Model with OpenStreetMap Data (NeurIPS 2025 · similarity 0.83)
- Spatial-Logic-Aware Weakly Supervised Learning for Flood Mapping on Earth Imagery (AAAI 2024 · similarity 0.83)
- Landsat30-AU: A Vision-Language Dataset for Australian Landsat Imagery (AAAI 2026 · 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.
