collaborative mapping, photography and 3d exploration of the urban environment

GeoSIM (Geo Social Image Mapping) is an urban sensing system with which a group of individuals with camera-equipped mobile phones participate in collaborative/social mapping of the urban image (i.e., the texture of the urban environment) at some target geolocation. The participating group, which may either consist of dedicated individuals or the general public, manually capture geotagged images of the urban environment. The collected images are progressively used for documentation of the dynamic urban image in multiple spatial resolutions at different times. We build GeoSIM based on GeoDec, a scalable geospatial data management engine that we have developed at USC. To develop GeoSIM, we extend GeoDec to GeoDec+ to support the requirements of GeoSIM, and develop GeoSIM as an application on top of GeoDec+ by introducing two novel enabling technologies with significant research challenges: 1) a coordination engine that plans participation of the GeoSIM users for efficient, fast and comprehensive social image mapping, while ensuring that the constraints and preferences of the participants are satisfied, and 2) a texture generation engine that depending on the available preexisting data (either low resolution texture together with the underlying geometric model,  only the geometric model, or no preexisting data at all), uses sophisticated  but efficient texture construction algorithms to progressively develop the urban image.