Unsupervised Learning for Improving Efficiency of Dense Three-Dimensional Scene Recovery in Corridor Mapping

  • Thomas Warsop
  • Sameer Singh
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6688)

Abstract

In this work, we perform three-dimensional scene recovery from image data capturing railway transportation corridors. Typical three-dimensional scene recovery methods initialise recovered feature positions by searching for correspondences between image frames. We intend to take advantage of a relationship between image data and recovered scene data to reduce the search space traversed when performing such correspondence matching. We build multi-dimensional Gaussian models of recurrent visual features associated with distributions representing recovery results from our own dense planar recovery method. Results show that such a scheme decreases the number of checks made per feature to 6% of a comparable exhaustive method, whilst unaffecting accuracy. Further, the proposed method performs competitively when compared with other methods presented in literature.

Keywords

Augmented Reality Image Frame Unsupervised Learn Epipolar Line British Machine Vision 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

References

  1. 1.
    Bay, H., Ess, A., Tuytelaars, T., Van Gool, L.: Surf: Speeded up robust features. Computer Vision and Image Understanding (CVIU) 110(3), 346–359 (2008)CrossRefGoogle Scholar
  2. 2.
    Chekhlov, D., Gee, A.P., Calway, A., Mayol-Cuevas, W.: Ninja on a plane: Automatic discovery of physical planes for augmented reality using visual slam. In: International Symposium on Mixed and Augmented Reality, ISMAR ( November 2007)Google Scholar
  3. 3.
    Chekhlov, D., Mayol-Cuevas, W.: Appearance based indexing for relocalisation in real-time visual slam. In: 19th British Machine Vision Conference, pp. 363–372 (2008)Google Scholar
  4. 4.
    Davison, A.J.: Real-time simulataneous localization and mapping with a single camera. In: Proc. International Conference on Computer Vision, pp. 1403–1411 (October 2003)Google Scholar
  5. 5.
    Davison, A.J., Reid, I.D., Molton, N.D., Stasse, O.: Monoslam: Real-time single camera slam. IEEE Transactions on PAtterns Analysis and Machine Intelligence 29, 1–15 (2007)CrossRefGoogle Scholar
  6. 6.
    Engels, C., Fraundorfer, F., Nistér, D.: Integration of tracked and recognized features for locally and globally robust structure from motion. In: VISAPP International Workshop on Robotic Perception, VISAPP RoboPerc (January 2008)Google Scholar
  7. 7.
    Favaro, P., Jin, H., Soatto, S.: A semi-direct approach to structure from motion. The Visual Computer 19, 377–384 (2003)CrossRefGoogle Scholar
  8. 8.
    Fraundorfer, F., Schindler, K., Bischof, H.: Piecewise planar scene reconstruction from sparse correspondences. In: Image and Vision Computing, vol. 24, pp. 395–406 (2006)Google Scholar
  9. 9.
    Furukawa, Y., Curless, B., Seitz, S.M., Szeliski, R.: Manhattan-world stereo. In: CVPR, pp. 1422–1429. IEEE, Los Alamitos (2009)Google Scholar
  10. 10.
    Gee, A.P., Chekhlov, D., Calway, A., Mayol-Cuevas, W.: Discovering higher level structure in visual slam. IEEE Transactions on Robotics 24, 980–990 (2008)CrossRefGoogle Scholar
  11. 11.
    Gee, A.P., Chekhlov, D., Mayol, W., Calway, A.: Discovering planes and collapsing the state space in visual slam. In: 18th British Machine Vision Conference ( September 2007)Google Scholar
  12. 12.
    Goecke, R., Asthana, A., Pettersson, N., Petersson, L.: Visual vehicle egomotion esitmation using the fourier-mellin transform. In: IEEE Intelligent Vehicles Symposium, pp. 450–455 (June 2007)Google Scholar
  13. 13.
    Kanbara, M., Ukita, N., Kidode, M., Yokoya, N.: 3d scene reconstruction from reflection images in a sphereical mirror. In: The 18th International Conference on Pattern Recognition (ICPR 2006), pp. 874–879 (2006)Google Scholar
  14. 14.
    Klein, G., Murray, D.: Parallel tracking and mapping for small ar workspaces. In: Proceedings and the Sixth IEEE and ACM International Symposium on Mixed and Augmented Reailty, ISMAR 2007 (2007)Google Scholar
  15. 15.
    Li, P., Farin, D., Gunnewiek, R.K., de With, P.H.N.: On creating depth maps from monoscopic video using structure frm motion. In: 27th Symposium on Information Theory, pp. 508–515 (2006)Google Scholar
  16. 16.
    Liu, G.-H., Feng, Q.-Y.: Recovering 3d shape and motion from image sequences using affine approximation. In: Second International Conference on Information and Computing Science, pp. 349–352 (2009)Google Scholar
  17. 17.
    Lowe, D.G.: Distinctive image features from scale-invariant keypoints. International Journal of Computer Vision (2004)Google Scholar
  18. 18.
    Martinez-Carranza, J., Calway, A.: Unifying planar and point mapping in monocular slam. In: British Machine Vision Conference, pp. 1–11 (September 2010)Google Scholar
  19. 19.
    Martinez-Carranza, J., Calway, A.: Efficiently increasing map density in visual slam using planar features with adaptive measurements. In: British Machine Vision Conference, pp. 1–11 ( September 2009)Google Scholar
  20. 20.
    Metro Solutions (2006), http://metrosolutions.org/go/doc/1068/116948 (last accessed: July 25, 2010)
  21. 21.
    Molton, N., Davidson, A., Reid, I.: Locally planar patch features for real-time structure from motion. In: Proc. British Machine Vision Conference, BMVC (September 2004)Google Scholar
  22. 22.
    Ogale, A.S., Aloimonos, Y.: Shape and the stereo correspondence problem. International Journal of Computer Vision 65, 147–162 ( December 2005)CrossRefGoogle Scholar
  23. 23.
    Pollefeys, M., Nister, D., Frahm, J.M., Akbarzadeh, A., Mordohai, P., Clipp, B., Engels, C., Gallup, D., Kim, S.J., Merrell, P., Salmi, C., Sinha, S.N., Talton, B., Wang, L., Yang, Q., Stewenius, H., Yang, R., Welch, G., Towles, H.: Detailed real-time urban 3d reconstruction from video. International Journal of Computer Vision, 143–167 (2008)Google Scholar
  24. 24.
    Sinha, S.N., Steedly, D., Szeliski, R.: Piecewise planar stereo for image-based rendering. In: Twelth IEEE International Conference on Computer Vision, ICCV 2009 (2009)Google Scholar
  25. 25.
    Tomasi, C., Kanade, T.: Shape and motion from image streams: a factorization method. Technical Report TR 92-1270, Carnegie Mellon (March 1992)Google Scholar
  26. 26.
    Yun, S.U., Min, D., Sohn, K.: 3d scene reconstruction system with hand-held stereo cameras. In: 3DTV Conference, pp. 1–4 (2007)Google Scholar
  27. 27.
    Zhang, G., Dong, Z., Jia, J., Wong, T.-T., Bao, H.: Efficient non-consecutive feature tracking for structure-from-motion. In: Daniilidis, K., Maragos, P., Paragios, N. (eds.) ECCV 2010. LNCS, vol. 6315, pp. 422–435. Springer, Heidelberg (2010)CrossRefGoogle Scholar
  28. 28.
    Zhang, G., Hua, W., Qin, X., Shao, Y., Bao, H.: Video stabilization based on a 3d perspective camera model. The Visual Computer 25, 997–1008 (2009)CrossRefGoogle Scholar
  29. 29.
    Zhang, G., Jia, J., Wong, T.-T., Bao, H.: Consistent depth maps recovery from a video sequence. IEEE Transactions on Pattern Analysis and Machine Intelligence 21, 974–988 (2009)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Thomas Warsop
    • 1
  • Sameer Singh
    • 1
  1. 1.Research School of Informatics, Holywell ParkLoughborough UniversityLeicestershireUK

Personalised recommendations