SSLChange: A Self-supervised Change Detection Framework Based on Domain Adaptation

TL;DR

SSLChange introduces a self-supervised contrastive learning framework for remote sensing change detection that reduces the need for manual labeling and enhances baseline performance, especially in data-limited scenarios.

Contribution

It proposes a novel self-supervised contrastive framework for change detection that can be pre-trained without labels and transferred to improve existing baselines.

Findings

Significant performance improvement in data-limited situations.

Enhanced stability of change detection baselines.

Effective transferability of the self-supervised model.

Abstract

In conventional remote sensing change detection (RS CD) procedures, extensive manual labeling for bi-temporal images is first required to maintain the performance of subsequent fully supervised training. However, pixel-level labeling for CD tasks is very complex and time-consuming. In this paper, we explore a novel self-supervised contrastive framework applicable to the RS CD task, which promotes the model to accurately capture spatial, structural, and semantic information through domain adapter and hierarchical contrastive head. The proposed SSLChange framework accomplishes self-learning only by taking a single-temporal sample and can be flexibly transferred to main-stream CD baselines. With self-supervised contrastive learning, feature representation pre-training can be performed directly based on the original data even without labeling. After a certain amount of labels are subsequently obtained, the pre-trained features will be aligned with the labels for fully supervised fine-tuning. Without introducing any additional data or labels, the performance of downstream baselines will experience a significant enhancement. Experimental results on 2 entire datasets and 6 diluted datasets show that our proposed SSLChange improves the performance and stability of CD baseline in data-limited situations. The code of SSLChange will be released at \url{https://github.com/MarsZhaoYT/SSLChange}