SatlasPretrain: A Large-Scale Dataset for Remote Sensing Image Understanding

TL;DR

SatlasPretrain introduces a large-scale, diverse remote sensing dataset with 302 million labels across 137 categories, enabling improved pre-training for various Earth monitoring tasks and highlighting the need for specialized methods.

Contribution

The paper presents SatlasPretrain, a comprehensive large-scale dataset for remote sensing, and demonstrates its effectiveness in enhancing model performance on downstream Earth observation tasks.

Findings

Pre-training on SatlasPretrain boosts accuracy by 18% over ImageNet.

Significant room for improvement exists in processing diverse remote sensing data.

Pre-trained models on SatlasPretrain outperform baselines on multiple tasks.

Abstract

Remote sensing images are useful for a wide variety of planet monitoring applications, from tracking deforestation to tackling illegal fishing. The Earth is extremely diverse -- the amount of potential tasks in remote sensing images is massive, and the sizes of features range from several kilometers to just tens of centimeters. However, creating generalizable computer vision methods is a challenge in part due to the lack of a large-scale dataset that captures these diverse features for many tasks. In this paper, we present SatlasPretrain, a remote sensing dataset that is large in both breadth and scale, combining Sentinel-2 and NAIP images with 302M labels under 137 categories and seven label types. We evaluate eight baselines and a proposed method on SatlasPretrain, and find that there is substantial room for improvement in addressing research challenges specific to remote sensing, including processing image time series that consist of images from very different types of sensors, and taking advantage of long-range spatial context. Moreover, we find that pre-training on SatlasPretrain substantially improves performance on downstream tasks, increasing average accuracy by 18% over ImageNet and 6% over the next best baseline. The dataset, pre-trained model weights, and code are available at https://satlas-pretrain.allen.ai/.