Clemens Rabe, Christoph Volmer, Uwe FrankeAbstractThis paper presents a system for monocular obstacle and lane boundary detection running in real-time. A Kalman Filter based depth from motion algorithm is used for the reconstruction of the three-dimensional scene. Using multiple filters in parallel the rate of convergence is significantly higher than in direct methods, especially if the vehicle drives slowly. In addition a pitch correction is introduced which improves the overall estimation in typical road scenarios. Real world examples illustrate the results of the proposed system. [Download] [Anzeigen] [BibTeX] |
Publikationen > 2005 >