Weakly-Supervised Domain Adaptation With Adversarial Entropy for Building Segmentation in Cross-Domain Aerial Imagery

• Main Authors: Xuedong Yao, Yandong Wang, Yanlan Wu, Zeyu Liang
• Format: Article
• Language: English
• Published: IEEE 2021-01-01
• Series: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
• Subjects: Adversarial learning; building segmentation; domain adaptation (DA); high resolution; self-training
• Online Access: https://ieeexplore.ieee.org/document/9517002/

Building segmentation is a classical and challenging task in high-resolution remote sensing imagery. This approach has achieved remarkable performance based on a fully convolutional network with adequate pixel-wise annotations. However, due to differences in sensor technology as well as appearance in different regions, datasets gathered from these various sources are quite distinct, and dense annotations for a particular area are not always available. Thus, directly applying a segmentation model trained on one dataset (source domain) to another unseen dataset (target domain) usually results in a drop in performance, called the domain gap. In this article, we propose a weakly-supervised domain adaptation method using adversarial entropy for building segmentation to address this problem. First, we use an adversarial entropy strategy to decrease the entropy and improve the prediction certainty for target images, causing the distributions between the source and target domains to become closer to each other. Second, we propose a simple and effective self-training strategy for the target domain that produces high-confidence predictions using pseudolabels. We use a series of thresholds to generate the pseudolabels without introducing extra parameters. This strategy effectively enhances the discriminability of the target domain and further minimizes the distribution discrepancy between the two domains. Experiments on cross-domain aerial datasets have demonstrated the effectiveness and superiority of our proposed method when compared to other state-of-the-art methods.