 This paper proposes a novel approach to building extraction from very high resolution, VHR, remote sensing images. It combines the advantages of deep learning and guided filtering to achieve better results than existing methods. The authors use a deep residual network to train their model on VHR remote sensing data, then employ a guided filter to further refine the output of the deep learning model. This produces a more accurate segmentation map of buildings in the urban area. Experiments show that the proposed method achieves significantly improved results over other state-of-the-art methods. This article was authored by Yongyang Su, Liangwu, Zhongxia, and others.