Simultaneous Localization and Mapping (SLAM)

A Stochastic Parallel Method for Real Time Monocular SLAM applied to Augmented Reality

Responsible:

Borro, Diego

Participants:

Sanchez, Jairo

ABSTRACT

In augmented reality applications, the position and orientation of the observer must be estimated in order to create a virtual camera that renders virtual objects aligned with the real scene. There are a wide variety of motion sensors available in the market, however, these sensors are usually expensive and impractical. In contrast, computer vision techniques can be used to estimate the camera pose using only the images provided by a single camera if the 3D structure of the captured scene is known beforehand. When it is unknown, some solutions use external markers, however, they require to modify the scene, which is not always possible.

Simultaneous Localization and Mapping (SLAM) techniques can deal with completely unknown scenes, simultaneously estimating the camera pose and the 3D structure. Traditionally, this problem is solved using nonlinear minimization techniques that are very accurate but hardly used in real time. In this way, this thesis presents a highly parallelizable random sampling approach based on Monte Carlo simulations that fits very well on the graphics hardware. As demonstrated in the text, the proposed algorithm achieves the same precision as nonlinear optimization, getting real time performance running on commodity graphics hardware.

PAPERS

Books:

  • Sánchez, J., A Stochastic Parallel Method for Real Time Monocular SLAM applied to Augmented Reality, PhD Thesis Dissertation in December 2010. (EG Digital Library pdf).

Journals:

    • Sánchez, J., Álvarez, H., and Borro, D., "GFT: GPU Fast Triangulation of 3D Points", Lecture Notes in Computer Science, Computer Vision and Graphics (ISBN-10: 3-642-15906-0), Vol. 6375, pp. 235-242. Ed: Springer-Verlag Berlin Heidelberg, New York. September 2010. (Book URL).

Conferences:

  • Sánchez, J., Álvarez, H., and Borro, D., "Towards Real time 3D Tracking and Reconstruction on a GPU using Monte Carlo Simulations", Proceedings of the 9th IEEE International Symposium on Mixed and Augmented Reality (ISMAR 2010), pp. 185-192 (ISBN: ISBN: 978-1-4244-9343-2). Coex, Seoul, Korea. October 13-16, 2010. (pdf, video).

  • Sánchez, J., Álvarez, H., and Borro, D., "GFT: GPU Fast Triangulation of 3D Points", Proceedings of the International Conference on Computer Vision and Graphics (ICCVG'10). Warsaw, Poland. September 20-22, 2010. (pdf, video).

  • Sánchez, J, and Borro, D., "Automatic Affine Structure Recovery Using RANSAC", Proceedings of the XX Congreso Español de Informática Gráfica (CEIG'10). Valencia, Spain. September 7-10, 2010. (pdf).

  • Eskudero, I., Sánchez, J., Buchart, C., García-Alonso, A., and Borro, D., “Tracking 3D en GPU basado en el Filtro de Partículas”, Proceedings of the XIX Congreso Español de Informática Gráfica (CEIG'09), pp. 47-55. San Sebastián, Spain. September 9-11, 2009. (pdf, video).

  • Sánchez, J., and Borro, D., "Non Invasive 3D Tracking for Augmented Video Applications", Proceedings of the IEEE Virtual Reality 2007 Conference, Workshop "Trends and Issues in Tracking for Virtual Environments", pp. 22-27. Charlotte, North Carolina, USA. March 10-14, 2007. (pdf).

Posters:

  • Sánchez, J., Álvarez, H., and Borro, D., "GPU Optimizer: a 3D Reconstruction on the GPU using Monte Carlo Simulations", Poster Proceedings of the International Conference on Computer Vision Theory and Applications (VISAPP'10), pp. 443-446. Angers, France. May 17-21, 2010. (pdf).

  • Sánchez, J., and Borro, D., "Automatic Augmented Video Creation for Markerless Environments", Poster Proceedings of the 2nd International Conference on Computer Vision Theory and Applications (VISAPP'07), pp. 519-522. Barcelona, Spain. March 8-11, 2007. (pdf).

VIDEOS

RELATED PROJECTS