Weakly Supervised Change Detection via Knowledge Distillation and Multiscale Sigmoid Inference

TL;DR

This paper introduces a weakly supervised change detection method that uses knowledge distillation and multiscale sigmoid inference to accurately identify changes from image-level labels, reducing the need for pixel-level annotations.

Contribution

The paper proposes a novel weakly supervised change detection approach combining knowledge distillation with multiscale sigmoid inference, improving accuracy without pixel-level labels.

Findings

Outperforms state-of-the-art on three public datasets

Effectively leverages image-level labels for change detection

Refines change maps with multiscale sigmoid inference

Abstract

Change detection, which aims to detect spatial changes from a pair of multi-temporal images due to natural or man-made causes, has been widely applied in remote sensing, disaster management, urban management, etc. Most existing change detection approaches, however, are fully supervised and require labor-intensive pixel-level labels. To address this, we develop a novel weakly supervised change detection technique via Knowledge Distillation and Multiscale Sigmoid Inference (KD-MSI) that leverages image-level labels. In our approach, the Class Activation Maps (CAM) are utilized not only to derive a change probability map but also to serve as a foundation for the knowledge distillation process. This is done through a joint training strategy of the teacher and student networks, enabling the student network to highlight potential change areas more accurately than teacher network based on image-level labels. Moreover, we designed a Multiscale Sigmoid Inference (MSI) module as a post processing step to further refine the change probability map from the trained student network. Empirical results on three public datasets, i.e., WHU-CD, DSIFN-CD, and LEVIR-CD, demonstrate that our proposed technique, with its integrated training strategy, significantly outperforms the state-of-the-art.