Self-Pair: Synthesizing Changes from Single Source for Object Change Detection in Remote Sensing Imagery

TL;DR

This paper introduces Self-Pair, a method that synthesizes changes from a single source image to improve change detection in remote sensing, addressing the challenge of limited bi-temporal training data.

Contribution

The paper proposes a novel approach that manipulates source images to better simulate changes, enhancing change detection accuracy with single-temporal supervision.

Findings

Outperforms existing single-temporal supervision methods

Maintains visual similarity in unchanged areas

Demonstrates effectiveness through extensive experiments

Abstract

For change detection in remote sensing, constructing a training dataset for deep learning models is difficult due to the requirements of bi-temporal supervision. To overcome this issue, single-temporal supervision which treats change labels as the difference of two semantic masks has been proposed. This novel method trains a change detector using two spatially unrelated images with corresponding semantic labels such as building. However, training on unpaired datasets could confuse the change detector in the case of pixels that are labeled unchanged but are visually significantly different. In order to maintain the visual similarity in unchanged area, in this paper, we emphasize that the change originates from the source image and show that manipulating the source image as an after-image is crucial to the performance of change detection. Extensive experiments demonstrate the importance of maintaining visual information between pre- and post-event images, and our method outperforms existing methods based on single-temporal supervision. code is available at https://github.com/seominseok0429/Self-Pair-for-Change-Detection.