Context Matters: Leveraging Spatiotemporal Metadata for Semi-Supervised Learning on Remote Sensing Images

TL;DR

This paper introduces a semi-supervised learning framework that leverages spatiotemporal metadata to improve pseudo-label quality in remote sensing images, enhancing model robustness to spatiotemporal shifts.

Contribution

It proposes a teacher-student SSL framework utilizing spatiotemporal metadata for pseudo-label refinement, with methods for encoding this information and a novel distillation mechanism.

Findings

Spatiotemporal metadata improves pseudo-label quality.

The teacher network benefits from metadata, while the student remains invariant at test time.

The framework enhances semi-supervised learning performance on remote sensing data.

Abstract

Remote sensing projects typically generate large amounts of imagery that can be used to train powerful deep neural networks. However, the amount of labeled images is often small, as remote sensing applications generally require expert labelers. Thus, semi-supervised learning (SSL), i.e., learning with a small pool of labeled and a larger pool of unlabeled data, is particularly useful in this domain. Current SSL approaches generate pseudo-labels from model predictions for unlabeled samples. As the quality of these pseudo-labels is crucial for performance, utilizing additional information to improve pseudo-label quality yields a promising direction. For remote sensing images, geolocation and recording time are generally available and provide a valuable source of information as semantic concepts, such as land cover, are highly dependent on spatiotemporal context, e.g., due to seasonal effects and vegetation zones. In this paper, we propose to exploit spatiotemporal metainformation in SSL to improve the quality of pseudo-labels and, therefore, the final model performance. We show that directly adding the available metadata to the input of the predictor at test time degenerates the prediction quality for metadata outside the spatiotemporal distribution of the training set. Thus, we propose a teacher-student SSL framework where only the teacher network uses metainformation to improve the quality of pseudo-labels on the training set. Correspondingly, our student network benefits from the improved pseudo-labels but does not receive metadata as input, making it invariant to spatiotemporal shifts at test time. Furthermore, we propose methods for encoding and injecting spatiotemporal information into the model and introduce a novel distillation mechanism to enhance the knowledge transfer between teacher and student. Our framework dubbed Spatiotemporal SSL can be easily combined with several stat...