Webseite Institut für Kraftfahrzeuge (ika)

Topic and Goals of the Thesis
For the development and safety validation of automated vehicles, huge trajectory datasets of road users are necessary. One solution to create those datasets is the use of drone imagery of traffic on e.g. highways or urban intersections.One challenge in producing aerial traffic recordings is that the drone inevitably moves or rotates away from the correct recording position due to wind and GPS drift. These movements have to be eliminated from the footage after the recording has been completed by stabilizing each video frame to the first image using salient image locations. A major difficulty, however, is that this must be done pixel-exactly for a long recording, since the moving vehicles themselves generate salient image locations and obscure the correct salient locations. The goal of this thesis is therefore to design, implement and evaluate a system for these special challenges.

Tasks
• Literature research on approaches on video stabilization and registration
• Design and implementation of a solution suitable for drone imagery
• Evaluation of the solution on a large amount of existing footage

Your Profile
• Good English and/or German language skills
• Reliability, commitment and enjoyment of working independently
• Basic computer vision experience
• Advanced programming experience in Python and/or Matlab

Um sich für diesen Job zu bewerben, sende deine Unterlagen per E-Mail an robert.krajewski@ika.rwth-aachen.de