Semisupervised Edge-Aware Road Extraction via Cross Teaching Between CNN and Transformer
Zi-Xiong Yang, Zhi-Hui You, Si-Bao Chen, Jin Tang, Bin Luo
Abstract
Semi-supervised semantic segmentation of remote sensing images has been proven to be an effective approach to reduce manual annotation costs and leverage available unlabeled data to improve segmentation performance. However, some existing methods that focus on self-training and consistent regularization fail to consider large-scale characteristics of remote sensing images and the importance of incorporating road edge information. In this paper, we propose a novel Semi-Supervised Edge-Aware Network (SSEANet) for remote sensing image semantic segmentation by jointly training CNN and transformer. SSEANet focuses on the consistency loss of multi-scale features and uses attention mechanism to fuse road edge information. Extensive experiments on DeepGlobe, Massachusetts and AerialKITTI-Bavaria datasets show that the proposed method outperforms state-of-the-arts, demonstrating its effectiveness.