Realtime 3d Mapping, Optimization, and Rendering Based on a Depth Sensor

Abstract

This thesis provides a method for using a portable scanner to create an optimized 3D map for real time rendering. This thesis uses a cloud computing software as a service architecture which allows for a portable scanner to acquire depth maps. Using a portable scanner allows for the mapping of large areas. It will then send the depth maps to a server for a 3D map to be created in real time. This thesis discusses the acquisition of the point cloud using an open source program for 3D mapping with a depth sensor. It then covers the triangulation into a mesh using the marching cubes algorithm. The optimization of the mesh to allow real time rendering is then introduced. Finally the mesh is imported and rendered in the Unreal Engine 3 for an interactive and intuitive display.