EasySeg: An Error-Aware Domain Adaptation Framework for Remote Sensing Imagery Semantic Segmentation via Interactive Learning and Active Learning

Semantic segmentation of remote sensing images has attracted much attention for its wide applications. While deep learning models have shown impressive performance in this task, challenges arise when applying them to data from other domains without fine-tuning, due to domain gaps. Domain adaptation (DA) has emerged as a solution to bridge this gap. Existing works mainly focus on unsupervised DA (UDA), which lags far behind fully supervised models. However, active DA (ADA) methods focused on natural images face challenges when applied to remote sensing images due to pronounced domain gaps and error unawareness problems. In this work, we propose a novel error-aware DA framework for remote sensing imagery semantic segmentation, called EasySeg, via interactive learning and active learning. First, we introduce a point-level labeling strategy, named “See-First-Ask-Later” (SFAL), combining both interactive and active learning manners, allowing obvious errors and information-rich pixels to be annotated easily and efficiently. Then, we introduce an interactive semantic segmentation network (ISS-Net), which can perform automatic semantic segmentation and interactive refinement. Based on the acquired target point-level labels, ISS-Net generates dense and accurate pseudo-labels to enhance DA performance through retraining with consistency regularization. Comprehensive experiments on two tasks demonstrate that our method outperforms the state-of-the-art ADA methods in terms of overall accuracy (OA), mean accuracy (MA),
$F1$
score, and mean intersection of union (mIoU) with lower labeling costs, even surpassing some fully supervised models. The source code of EasySeg is freely available at
https://github.com/Yangliangzhe/EasySeg
.