Despite recent progress in computer vision, fine-grained interpretation of satellite images remains challenging because of a lack of labeled training data. To overcome this limitation, we propose using Wikipedia as a previously untapped source of rich, georeferenced textual information with global coverage. We construct a novel large-scale, multi-modal dataset by pairing geo-referenced Wikipedia articles with satellite imagery of their corresponding locations. To prove the efficacy of this dataset, we focus on the African continent and train a deep network to classify images based on labels extracted from articles. We then fine-tune the model on a human annotated dataset and demonstrate that this weak form of supervision can drastically reduce the quantity of human annotated labels and time required for downstream tasks.
@article{arxiv.1809.10236,
title = {Learning to Interpret Satellite Images Using Wikipedia},
author = {Evan Sheehan and Burak Uzkent and Chenlin Meng and Zhongyi Tang and Marshall Burke and David Lobell and Stefano Ermon},
journal= {arXiv preprint arXiv:1809.10236},
year = {2018}
}