Generating a three-dimensional model from images captured by an unmanned aerial vehicle

This paper presents a method for three-dimensional reconstruction using only the images captured from an Unmanned Aerial Vehicle (UAV). First, the camera pose is computed using a known marker on the scene as reference, then, a three dimensional point cloud reconstruction is performed using Multi-View Stereo software (MVS); and finally the polygon mesh is generated using the Advancing Front technique. The results show that the obtained polygon mesh is a good approximation of the surface, precise enough for applying standard algorithms for path planning and obstacle avoidance.