Revisiting Consistency Regularization for Semi-supervised Change Detection in Remote Sensing Images

TL;DR

This paper introduces a semi-supervised change detection method for remote sensing images that leverages unlabeled data through consistency regularization, reducing the need for extensive annotations while maintaining high performance.

Contribution

It proposes a novel semi-supervised change detection model that enforces output consistency under perturbations, improving performance with limited labeled data.

Findings

Achieves near-supervised performance with only 10% labeled data.

Effectively leverages unlabeled data to improve change detection accuracy.

Demonstrates robustness of the method across multiple datasets.

Abstract

Remote-sensing (RS) Change Detection (CD) aims to detect "changes of interest" from co-registered bi-temporal images. The performance of existing deep supervised CD methods is attributed to the large amounts of annotated data used to train the networks. However, annotating large amounts of remote sensing images is labor-intensive and expensive, particularly with bi-temporal images, as it requires pixel-wise comparisons by a human expert. On the other hand, we often have access to unlimited unlabeled multi-temporal RS imagery thanks to ever-increasing earth observation programs. In this paper, we propose a simple yet effective way to leverage the information from unlabeled bi-temporal images to improve the performance of CD approaches. More specifically, we propose a semi-supervised CD model in which we formulate an unsupervised CD loss in addition to the supervised Cross-Entropy (CE) loss by constraining the output change probability map of a given unlabeled bi-temporal image pair to be consistent under the small random perturbations applied on the deep feature difference map that is obtained by subtracting their latent feature representations. Experiments conducted on two publicly available CD datasets show that the proposed semi-supervised CD method can reach closer to the performance of supervised CD even with access to as little as 10% of the annotated training data. Code available at https://github.com/wgcban/SemiCD