AI Paper Atlas

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

ICCV 2025 · Vision-language and multimodal › Human and embodied modeling · approach: Medical vision-language models · 54 citations (OpenAlex)

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 changedWhereMeasuredEffect
add metadata encoding (geolocation, area, time) (ablation)Copernicus-FM · EuroSAT-S1OA: 56.9 -> 81.0Improved
add metadata encoding (ablation)Copernicus-FM · AQ-O3-S5PRMSE: 1556.3 -> 811.6Improved
add metadata encoding (ablation)Copernicus-FM · EuroSAT-S2OA: 88.9 -> 88.9No clear effect
add Fourier bandwidth encoding to the spectral hypernetwork (ablation)Copernicus-FM · EuroSAT-RGBOA: 62.2 -> 65.4Improved
add Fourier bandwidth encoding to the spectral hypernetwork (ablation)Copernicus-FM · AQ-O3-S5PRMSE: 2218.0 -> 1710.7Improved
add variable hypernetwork (ablation)Copernicus-FM · LC100-S3mAP: 87.1 -> 86.6Hurt
add variable hypernetwork (ablation)Copernicus-FM · EuroSAT-S1OA: 78.9 -> 81.0Improved
add continual distillation (ablation)Copernicus-FM · EuroSAT-S1OA: 77.9 -> 81.0Improved
add continual distillation (ablation)Copernicus-FM · LC100-S3mAP: 90.7 -> 90.7No clear effect
use LLM variable name encoding instead of random hash (ablation)Copernicus-FM · EuroSAT-S1OA: 77.7 -> 81.0Improved
use LLM encoding instead of spectroscopy descriptions (ablation)Copernicus-FM · EuroSAT-S1OA: 80.4 -> 81.0Improved
raise metadata drop ratio from 0.3 to 0.7 (ablation)Copernicus-FM · EuroSAT-S1 (k-NN)OA: 73.7 -> 81.0Improved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy