Continuation of Comprehensive Conjoint GPS, Sensor and Video Data Analysis for Next Generation of Smart Maps

In recent years, the availability of spatial data has developed rapidly. Characteristic of this development is the involvement of a large number of users, often using smartphones and mobile devices, to generate and make freely available Volunteered Geographic Information (VGI). While GPS and gyroscope data (e.g. from fitness bands) are common, the huge amount of data that is easily and typically collected in the form of videos makes video analysis based methods very demanding. On the other hand, only videos allow a comprehensive interpretation of the scene. In this research project we are interested in combining GPS, gyroscope and video data to analyse road and traffic situations for cyclists and pedestrians. Our standard setup is a smartphone attached to a bicycle that records GPS coordinates, videos, (online) local weather information and time. We will (a) use the GPS data for integration into a map, (b) use local speed and gyroscope data as well as variations in the sensor data to identify important situations during a bike ride, and (c) use video data to understand these important situations, e.g. causing a delay in the ride.The analysed data can then be used for map enhancement and route recommendation, but also for identification of unclear road markings, which is important for urban planning and accident avoidance. In addition to gaining experience with real-time and event-driven data with a focus on human-centred applications, the foundations can easily be extended to traffic management after hazards, quality control of topographic data sets, or environmental and health-related data analysis using additional sensor information.


Prof. Dr.-Ing. Bodo Rosenhahn

Project Manager