Digital twin technology based on building a virtual digital city similar to a real one enables the simulation of urban phenomena or the design of a city. A geospatial platform is an essential supporting component of digital twin cities. In this study, we propose a planetary-scale geospatial open platform that can be used easily in the most widely used game engine environment. The proposed platform can visualize large-capacity geospatial data in real time because it organizes and manages various types of data based on quadtree tiles. The proposed rendering tile decision method provides constant geospatial data visualization according to the camera controls of the user. The platform implemented is based on Unity3D, and therefore, one can use it easily by importing the proposed asset library. The proposed geospatial platform is available on the Asset Store. We believe that the proposed platform can meet the needs of various three-dimensional (3-D) geospatial applications.

Density-based clustering algorithms have been the most commonly used algorithms for discovering regions and points of interest in cities using global positioning system (GPS) information in geo-tagged photos. However, users sometimes find more specific areas of interest using real objects captured in pictures. Recent advances in deep learning technology make it possible to recognize these objects in photos. However, since deep learning detection is a very time-consuming task, simply combining deep learning detection with density-based clustering is very costly. In this paper, we propose a novel algorithm supporting deep content and density-based clustering, called deep density-based spatial clustering of applications with noise (DeepDBSCAN). DeepDBSCAN incorporates object detection by deep learning into the density clustering algorithm using the nearest neighbor graph technique. Additionally, this supports a graph-based reduction algorithm that reduces the number of deep detections. We performed experiments with pictures shared by users on Flickr and compared the performance of multiple algorithms to demonstrate the excellence of the proposed algorithm.