Body-relative Navigation Using Uncalibrated Cameras
Olivier Koch
MIT, Computer Science and Artifical Intelligence Laboratory, Boston, MA, USA
Talk
Olivier Koch is a PhD student at the MIT Computer Science and Artificial Laboratory. His research interests include 2D/3D geometry as well as localization and navigation using vision. He holds an engineering degree from Ecole Nationale Superieure de Techniques Avancees (France) and a MSc degree from MIT. Olivier was a member of the Darpa Urban Challenge MIT Team in 2007.
URL: http://people.csail.mit.edu/koch
We present a vision-based method that assists human navigation within unfamiliar environments. Our main contribution is a novel algorithm that learns the correlation between user egomotion and feature matches on a wearable set of uncalibrated cameras. The primary advantage of this method is that it provides robust guidance cues in the user's body frame, and is tolerant to small changes in the camera configuration. We couple this method with a topological mapping algorithm that provides global localization within the traversed environment. We validate our approach with ground-truth experiments and demonstrate the method on several real-world datasets spanning two kilometers of indoor and outdoor walking excursions.