Research map / Vision-language and multimodal / Human and embodied modeling
Towards a Unified Copernicus Foundation Model for Earth Vision
Yi Wang, Zhitong Xiong, Chenying Liu, Adam J. Stewart, Thomas Dujardin, Nikolaos Ioannis Bountos, Angelos Zavras, Franziska Gerken, Ioannis Papoutsis, Laura Leal-Taixé, Xiao Xiang Zhu
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 metadata encoding (geolocation, area, time) (ablation) | Copernicus-FM · EuroSAT-S1 | OA: 56.9 -> 81.0 | Improved |
| add metadata encoding (ablation) | Copernicus-FM · AQ-O3-S5P | RMSE: 1556.3 -> 811.6 | Improved |
| add metadata encoding (ablation) | Copernicus-FM · EuroSAT-S2 | OA: 88.9 -> 88.9 | No clear effect |
| add Fourier bandwidth encoding to the spectral hypernetwork (ablation) | Copernicus-FM · EuroSAT-RGB | OA: 62.2 -> 65.4 | Improved |
| add Fourier bandwidth encoding to the spectral hypernetwork (ablation) | Copernicus-FM · AQ-O3-S5P | RMSE: 2218.0 -> 1710.7 | Improved |
| add variable hypernetwork (ablation) | Copernicus-FM · LC100-S3 | mAP: 87.1 -> 86.6 | Hurt |
| add variable hypernetwork (ablation) | Copernicus-FM · EuroSAT-S1 | OA: 78.9 -> 81.0 | Improved |
| add continual distillation (ablation) | Copernicus-FM · EuroSAT-S1 | OA: 77.9 -> 81.0 | Improved |
| add continual distillation (ablation) | Copernicus-FM · LC100-S3 | mAP: 90.7 -> 90.7 | No clear effect |
| use LLM variable name encoding instead of random hash (ablation) | Copernicus-FM · EuroSAT-S1 | OA: 77.7 -> 81.0 | Improved |
| use LLM encoding instead of spectroscopy descriptions (ablation) | Copernicus-FM · EuroSAT-S1 | OA: 80.4 -> 81.0 | Improved |
| raise metadata drop ratio from 0.3 to 0.7 (ablation) | Copernicus-FM · EuroSAT-S1 (k-NN) | OA: 73.7 -> 81.0 | Improved |
Closest papers by meaning
- GeoCore-9B: Towards Geo-Aware Generative Foundation Models in Earth Observation (NeurIPS 2026 · similarity 0.86)
- SkySense: A Multi-Modal Remote Sensing Foundation Model Towards Universal Interpretation for Earth Observation Imagery (CVPR 2024 · similarity 0.85)
- REOBench: Benchmarking Robustness of Earth Observation Foundation Models (NeurIPS 2025 · similarity 0.85)
- DiffusionSat: A Generative Foundation Model for Satellite Imagery (ICLR 2024 · similarity 0.85)
- Galileo: Learning Global & Local Features of Many Remote Sensing Modalities (ICML 2025 · similarity 0.85)
- AnySat: One Earth Observation Model for Many Resolutions, Scales, and Modalities (CVPR 2025 · similarity 0.85)
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.
