Outdoor Visual Topometric Localization 1

The video shows the results of the visual topometric localization [1] in an outdoor environment. The top two views show the current view of the vehicle. The views corresponding to the best matching location are shown at the bottom. The evaluation sequence was acquired on September 19, 2010. The learning was performed with a data set acquired on October 28, 2010. Both data sets were acquired while the vehicle travelled the same 8 km route in the city of Pittsburgh, PA. More details can be found here: [1] Hernan Badino, Daniel Huber and Takeo Kanade, "Visual Topometric Localization". In Intelligent Vehicles Symposium, Baden Baden, June, 2011. See also: Hernan Badino, Daniel Huber and Takeo Kanade, "Real-Time Topometric Localization". In International Conference on Robotics and Automation (ICRA12), St. Paul, Minnesota, USA, May, 2012. Hernan Badino http://www.lelaps.de/ Video was generated using QCV. QCV is a qt-based computer vision framework for Linux. It can be found at SourceForge.net http://sourceforge.net/projects/qcv/