PhilEO Bench: Evaluating Geo-Spatial Foundation Models
This provides a standardized benchmark for evaluating geo-spatial foundation models, addressing a bottleneck for researchers in remote sensing, though it is incremental as it builds on existing model evaluation approaches.
The paper tackles the lack of annotated data in Earth Observation by introducing PhilEO Bench, a 400 GB evaluation framework with labeled tasks, and shows results from experiments on models like Prithvi and SatMAE at various n-shots and convergence rates.
Massive amounts of unlabelled data are captured by Earth Observation (EO) satellites, with the Sentinel-2 constellation generating 1.6 TB of data daily. This makes Remote Sensing a data-rich domain well suited to Machine Learning (ML) solutions. However, a bottleneck in applying ML models to EO is the lack of annotated data as annotation is a labour-intensive and costly process. As a result, research in this domain has focused on Self-Supervised Learning and Foundation Model approaches. This paper addresses the need to evaluate different Foundation Models on a fair and uniform benchmark by introducing the PhilEO Bench, a novel evaluation framework for EO Foundation Models. The framework comprises of a testbed and a novel 400 GB Sentinel-2 dataset containing labels for three downstream tasks, building density estimation, road segmentation, and land cover classification. We present experiments using our framework evaluating different Foundation Models, including Prithvi and SatMAE, at multiple n-shots and convergence rates.