BARCELONA, Spain - Jakob Engel, PhD student in the Computer Vision Group at the Technical University of Munich, Germany, has received the EMVA Young Professional Award 2014. Engel received the award for his novel approach to real-time visual odometry for a monocular camera.
In his awarded work, Engel developed a novel, semi-dense method to estimate the 3D movement of a monocular camera from the recorded video and simultaneously estimating the 3D structure of the environment. The underlying task is called (monocular) visual odometry and is used, for example, in augmented- or virtual-reality applications or the autonomous navigation of robots. In contrast to the classical keypoint-based approach, the proposed semi-dense, direct formulation allows the utilization of all information in the video, achieving superior tracking accuracy and much more detailed reconstructions, while still running in real time, even on a modern smartphone.