In-Domain Self-Supervised Learning Improves Remote Sensing Image Scene Classification

TL;DR

This paper demonstrates that in-domain self-supervised pre-training using large remote sensing datasets significantly enhances the accuracy of remote sensing image scene classification compared to traditional methods.

Contribution

The study shows that employing in-domain SSL pre-training with Vision transformers on large remote sensing datasets improves downstream classification performance.

Findings

In-domain SSL pre-training outperforms generic pre-training methods.

Using the iBOT framework with Million-AID dataset yields better results.

Performance gains are consistent across 14 diverse datasets.

Abstract

We investigate the utility of in-domain self-supervised pre-training of vision models in the analysis of remote sensing imagery. Self-supervised learning (SSL) has emerged as a promising approach for remote sensing image classification due to its ability to exploit large amounts of unlabeled data. Unlike traditional supervised learning, SSL aims to learn representations of data without the need for explicit labels. This is achieved by formulating auxiliary tasks that can be used for pre-training models before fine-tuning them on a given downstream task. A common approach in practice to SSL pre-training is utilizing standard pre-training datasets, such as ImageNet. While relevant, such a general approach can have a sub-optimal influence on the downstream performance of models, especially on tasks from challenging domains such as remote sensing. In this paper, we analyze the effectiveness of SSL pre-training by employing the iBOT framework coupled with Vision transformers trained on Million-AID, a large and unlabeled remote sensing dataset. We present a comprehensive study of different self-supervised pre-training strategies and evaluate their effect across 14 downstream datasets with diverse properties. Our results demonstrate that leveraging large in-domain datasets for self-supervised pre-training consistently leads to improved predictive downstream performance, compared to the standard approaches found in practice.