AnySat: One Earth Observation Model for Many Resolutions, Scales, and Modalities
This addresses the challenge of handling varied resolutions, scales, and modalities in Earth observation for environmental monitoring applications, representing a novel method rather than an incremental improvement.
The authors tackled the problem of adapting geospatial models to diverse Earth observation data by proposing AnySat, a unified model that achieved state-of-the-art results on multiple datasets for tasks like land cover mapping and change detection.
Geospatial models must adapt to the diversity of Earth observation data in terms of resolutions, scales, and modalities. However, existing approaches expect fixed input configurations, which limits their practical applicability. We propose AnySat, a multimodal model based on joint embedding predictive architecture (JEPA) and scale-adaptive spatial encoders, allowing us to train a single model on highly heterogeneous data in a self-supervised manner. To demonstrate the advantages of this unified approach, we compile GeoPlex, a collection of 5 multimodal datasets with varying characteristics and $11$ distinct sensors. We then train a single powerful model on these diverse datasets simultaneously. Once fine-tuned or probed, we reach state-of-the-art results on the test sets of GeoPlex and for 6 external datasets across various environment monitoring tasks: land cover mapping, tree species identification, crop type classification, change detection, climate type classification, and segmentation of flood, burn scar, and deforestation. The code and models are available at https://github.com/gastruc/AnySat.