# Learning to Interpret Satellite Images in Global Scale Using Wikipedia

arXiv: https://arxiv.org/abs/1905.02506 · HTML: https://mlsanity.com/paper/1905.02506/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 7 May 2019.

## Built on by (most cited first, 8 of 8)

- [Seasonal Contrast: Unsupervised Pre-Training from Uncurated Remote Sensing Data](https://mlsanity.com/paper/2103.16607/) (2021)
- [Mission Critical -- Satellite Data is a Distinct Modality in Machine Learning](https://mlsanity.com/paper/2402.01444/) (2024)
- [Hard-Attention for Scalable Image Classification](https://mlsanity.com/paper/2102.10212/) (2021)
- [Mind the Modality Gap: Towards a Remote Sensing Vision-Language Model via Cross-modal Alignment](https://mlsanity.com/paper/2402.09816/) (2024)
- [Evaluating the Label Efficiency of Contrastive Self-Supervised Learning for Multi-Resolution Satellite Imagery](https://mlsanity.com/paper/2210.06786/) (2022)
- [TLDR: Text Based Last-layer Retraining for Debiasing Image Classifiers](https://mlsanity.com/paper/2311.18291/) (2023)
- [Predicting Poverty Level from Satellite Imagery using Deep Neural Networks](https://mlsanity.com/paper/2112.00011/) (2021)
- [Annotating Satellite Images of Forests with Keywords from a Specialized Corpus in the Context of Change Detection](https://mlsanity.com/paper/2509.13586/) (2025)

## Abstract (arXiv)

Despite recent progress in computer vision, finegrained interpretation of satellite images remains challenging because of a lack of labeled training data. To overcome this limitation, we construct a novel dataset called WikiSatNet by pairing georeferenced Wikipedia articles with satellite imagery of their corresponding locations. We then propose two strategies to learn representations of satellite images by predicting properties of the corresponding articles from the images. Leveraging this new multi-modal dataset, we can drastically reduce the quantity of human-annotated labels and time required for downstream tasks. On the recently released fMoW dataset, our pre-training strategies can boost the performance of a model pre-trained on ImageNet by up to 4:5% in F1 score.