The capability to directly access 3D content using browser is a chance given today by WebGL technologies. This allows to use the online globe together with your own data in any browser on any platform, mobile devices included. The purpose of this thesis is to implement a WebGL Geobrowser open source software enabling visualization of maps, satellite imagery and aerial photography on top of a virtual terrain, starting from existing OpenSource platform. In particular, the focus will be on the capabilities to access to typical OGC standards (WMS). The candidate will extend the existing functionalities (Globe rotation and zoom, camera’s tilt, free movement in space, displays existing maps) with new features such as: picking support to custom map tiles, open location services (OLS) etc.
Segmentation of building facades has become a central problem in recent computer vision. Many approaches for automatic 3D city reconstruction exist, but they are still missing an important feature, i.e. detailed facades. The aim of the present work will be to combine 3D acquisition techniques such as photogrammetric reconstructions with a procedural approach so as to create complex facade rule templates. The candidate will implement new rules in order to improve the quality of 3D city Models, using existing tools and 3D models generated using photogrammetric techniques.
Computer graphics and image processing are among the most important tools for geospatial applications, so it is no surprise that graphics algorithms and hardware have long played a vital role in geospatial computing. Modern GPUs have outgrown their graphics heritage in many ways to emerge as the world's most successful parallel computing architecture. Point cloud rendering is a crucial component for LIDAR data processing. GPGPU has great promise for bringing the distributing processes to the geospatial research and application, such as point clouds based LiDAR data processing and large remote sensing image processing, etc. The work aim is to improve the point cloud rendering using powerful GPUs for computing using a massive parallel approach.