SMAPGAN：基于生成式对抗网络的半监督样式地图瓦片生成方法
SMAPGAN: Generative Adversarial Network Based Semi-Supervised Styled Map Tiles Generating Method
摘要：
在互联网上广泛使用的传统在线地图瓦片，如谷歌地图和百度地图，都是从矢量数据中渲染出来的。从矢量数据中及时更新在线地图瓦片是一项困难的任务，其生成是耗时的。利用遥感影像及时生成地图分块是一条捷径，它可以在不需要矢量数据的情况下实时获取。然而，这个任务曾经是具有挑战性的，甚至是不可能的。受基于生成对抗网络（GAN）的图像到图像转换（img2img）技术的启发，提出了一种基于生成对抗网络（SMAPGAN）模型的半监督风格化地图瓦片生成方法，直接从遥感图像生成风格化地图瓦片。在这个模型中，我们设计了一个半监督学习策略，在丰富的未配对样本上对SMAPGAN进行预训练，并在现实中有限的配对样本上对其进行微调。我们还设计了图像梯度L1损失和图像梯度结构损失，以生成具有全局拓扑关系和详细的对象的边缘曲线，这是重要的制图风格的地图瓦片。此外，我们提出了边缘结构相似性指数（ESSI）作为一个度量来评估生成的地图瓦片和地面实况之间的拓扑一致性的质量。实验结果表明，SMAPGAN优于国家的最先进的（SOTA）作品根据均方误差，结构相似性指数，和ESSI。此外，SMAPGAN赢得了更多的认可比SOTA在人类感知测试的视觉现实主义的地图。我们的工作表明，SMAPGAN是一个潜在的新范式，以产生风格的地图瓦片。我们的SMAPGAN实现可在https://github.com/imcsq/SMAPGAN上获得。
英文摘要：
Traditional online map tiles, widely used on the Internet such as Google Map and Baidu Map, are rendered from vector data. Timely updating online map tiles from vector data, of which the generating is time-consuming, is a difficult mission. It is a shortcut to generate map tiles in time from remote sensing images, which can be acquired timely without vector data. However, this mission used to be challenging or even impossible. Inspired by image-to-image translation (img2img) techniques based on generative adversarial networks (GAN), we proposed a semi-supervised Generation of styled map Tiles based on Generative Adversarial Network (SMAPGAN) model to generate styled map tiles directly from remote sensing images. In this model, we designed a semi-supervised learning strategy to pre-train SMAPGAN on rich unpaired samples and fine-tune it on limited paired samples in reality. We also designed image gradient L1 loss and image gradient structure loss to generate a styled map tile with global topological relationships and detailed edge curves of objects, which are important in cartography. Moreover, we proposed edge structural similarity index (ESSI) as a metric to evaluate the quality of topological consistency between generated map tiles and ground truths. Experimental results present that SMAPGAN outperforms state-of-the-art (SOTA) works according to mean squared error, structural similarity index, and ESSI. Also, SMAPGAN won more approval than SOTA in the human perceptual test on the visual realism of cartography. Our work shows that SMAPGAN is potentially a new paradigm to produce styled map tiles. Our implementation of the SMAPGAN is available at https://github.com/imcsq/SMAPGAN.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
