Initialisation for Visual Tracking in Urban Environments

Outdoor augmented reality systems often rely on GPS to cover large environments. Visual tracking approaches can provide more accurate location estimates but typically require a manual initialisation procedure. This paper describes the combination of both techniques to create an accurate localisation system that does not require any additional input for (re-)initialisation. The 2D GPS position together with average user height is used as an initial estimate for the visual tracking. The large gap in available GPS accuracy versus required accuracy for initialisation is overcome through a search procedure that tries to minimise search time by improving the likelihood of finding the correct estimate early. Re-initialisation of the visual tracking system after catastrophic failures is further improved by modelling the GPS error with a Gaussian process to provide a better estimate of the current location, thereby decreasing search time.

[1]  Bruce H. Thomas,et al.  Tinmith-Metro: new outdoor techniques for creating city models with an augmented reality wearable computer , 2001, Proceedings Fifth International Symposium on Wearable Computers.

[2]  Tom Drummond,et al.  Going out: robust model-based tracking for outdoor augmented reality , 2006, 2006 IEEE/ACM International Symposium on Mixed and Augmented Reality.

[3]  Mark A. Sturza,et al.  GPS Navigation Using Three Satellites and a Precise Clock , 1983 .

[4]  Greg Welch,et al.  SCAAT: incremental tracking with incomplete information , 1997, SIGGRAPH.

[5]  Ronald Azuma,et al.  Orientation Tracking for Outdoor Augmented Reality Registration , 1999, IEEE Computer Graphics and Applications.

[6]  V. Renaudin,et al.  Hybridization of MEMS and Assisted GPS for Pedestrian Navigation , 2007 .

[7]  Roberto Cipolla,et al.  Visual tracking and control using Lie algebras , 1999, Proceedings. 1999 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Cat. No PR00149).

[8]  Eric Foxlin,et al.  Pedestrian tracking with shoe-mounted inertial sensors , 2005, IEEE Computer Graphics and Applications.

[9]  Shuzhi Sam Ge,et al.  Autonomous vehicle positioning with GPS in urban canyon environments , 2001, IEEE Trans. Robotics Autom..

[10]  Steven K. Feiner,et al.  Exploring MARS: developing indoor and outdoor user interfaces to a mobile augmented reality system , 1999, Comput. Graph..

[11]  Alice M. Agogino,et al.  Sensor noise model development of a longitudinal positioning system for AVCS , 1999, Proceedings of the 1999 American Control Conference (Cat. No. 99CH36251).

[12]  Roberto Cipolla,et al.  An Image-Based System for Urban Navigation , 2004, BMVC.

[13]  T. Drummond,et al.  Going out : Robust Tracking for Outdoor Augmented Reality , 2006 .

[14]  Amine Bermak,et al.  Gaussian process for nonstationary time series prediction , 2004, Comput. Stat. Data Anal..

[15]  Gérard Lachapelle,et al.  GNSS Indoor Location Technologies , 2004 .

[16]  Volker Coors,et al.  Matching buildings: pose estimation in an urban environment , 2000, Proceedings IEEE and ACM International Symposium on Augmented Reality (ISAR 2000).

[17]  David J. C. MacKay,et al.  Information Theory, Inference, and Learning Algorithms , 2004, IEEE Transactions on Information Theory.

[18]  J. D. Weiss,et al.  GPS/INS integration in a severe urban environment , 1998, IEEE 1998 Position Location and Navigation Symposium (Cat. No.98CH36153).