RANGE: Retrieval Augmented Neural Fields for Multi-Resolution Geo-Embeddings
It addresses the need for better visual feature capture in geospatial embeddings, benefiting tasks like species classification and population estimation, but is incremental as it builds on prior contrastive alignment methods.
The paper tackles the problem of geographic location representation for geospatial tasks by proposing RANGE, a retrieval-augmented strategy that improves over existing methods, achieving gains of up to 13.1% on classification and 0.145 R² on regression tasks.
The choice of representation for geographic location significantly impacts the accuracy of models for a broad range of geospatial tasks, including fine-grained species classification, population density estimation, and biome classification. Recent works like SatCLIP and GeoCLIP learn such representations by contrastively aligning geolocation with co-located images. While these methods work exceptionally well, in this paper, we posit that the current training strategies fail to fully capture the important visual features. We provide an information-theoretic perspective on why the resulting embeddings from these methods discard crucial visual information that is important for many downstream tasks. To solve this problem, we propose a novel retrieval-augmented strategy called RANGE. We build our method on the intuition that the visual features of a location can be estimated by combining the visual features from multiple similar-looking locations. We evaluate our method across a wide variety of tasks. Our results show that RANGE outperforms the existing state-of-the-art models with significant margins in most tasks. We show gains of up to 13.1% on classification tasks and 0.145 $R^2$ on regression tasks. All our code and models will be made available at: https://github.com/mvrl/RANGE.