Demo

Abstract

3D maps of urban environments are useful in various fields, from cellular network planning to urban planning and climatology. We show that 3D urban maps can be extracted from received satellite signals that are attenuated by obstacles, such as buildings, from low-accuracy GNSS data, crowdsourced opportunistically from standard smartphones during their user's uncontrolled daily commute trips. Our proposal incorporates position inaccuracies in the calculations, and the diversity of collection conditions of crowdsourced GNSS positions is used to mitigate bias and noise from the data. A binary classification model is trained and evaluated on multiple urban scenarios. Our results show that the generalization accuracy for a Random Forest classifier lies between 79% and 91%, demonstrating the potential of the proposed method for building 3D maps for wide urban areas. In the demo, we show multiple 3D visualizations of the various processing stages, which can be viewed interactively using Google Earth, allowing hands-on exploration of the work.