International Journal of Computer Vision

, Volume 98, Issue 2, pp 202–216 | Cite as

Egomotion Estimation Using Assorted Features

Article

Abstract

We propose a novel minimal solver for recovering camera motion across two views of a calibrated stereo rig. The algorithm can handle any assorted combination of point and line features across the four images and facilitates a visual odometry pipeline that is enhanced by well-localized and reliably-tracked line features while retaining the well-known advantages of point features. The mathematical framework of our method is based on trifocal tensor geometry and a quaternion representation of rotation matrices. A simple polynomial system is developed from which camera motion parameters may be extracted more robustly in the presence of severe noise, as compared to the conventionally employed direct linear/subspace solutions. This is demonstrated with extensive experiments and comparisons against the 3-point and line-sfm algorithms.

Keywords

Visual odometry SLAM Structure from motion Tracking 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Ansar, A., & Daniilidis, K. (2003). Linear pose estimation from points or lines. IEEE Transactions on Pattern Analysis and Machine Intelligence, 25(5), 578–589. CrossRefGoogle Scholar
  2. Bartoli, A., & Sturm, P. (2003). Multiple-view structure and motion from line correspondences. In ICCV. Google Scholar
  3. Bujnak, M., Kukelova, Z., & Pajdla, T. (2008). A general solution to the p4p problem for camera with unknown focal length. In IEEE computer society conference on computer vision and pattern recognition (CVPR) (pp. 1–8). Google Scholar
  4. Chandraker, M., Lim, J., & Kreigman, D. J. (2009). Moving in stereo: efficient structure and motion using lines. In ICCV. Google Scholar
  5. Christy, S., & Horaud, R. (1999). Iterative pose computation from line correspondences. Computer Vision and Image Understanding, 73(1), 137–144. MATHCrossRefGoogle Scholar
  6. Comport, A., Malis, E., & Rives, P. (2007). Accurate quadrifocal tracking for robust 3d visual odometry. In ICRA (pp. 40–45). Google Scholar
  7. Dornaika, F., & Garcia, C. (1999). Pose estimation using point and line correspondences. Real-Time Imaging, 5(3), 215–230. CrossRefGoogle Scholar
  8. Fischler, M. A., & Bolles, R. C. (1997). Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography. International Journal of Computer Vision, 22(2), 125–140. CrossRefGoogle Scholar
  9. Haralick, R., Lee, C., Ottenberg, K., & Nolle, M. (1991). Analysis and solutions of the three point perspective pose estimation problem. In IEEE computer society conference on computer vision and pattern recognition (CVPR). Google Scholar
  10. Hartley, R. (1997). Lines and points in three views and the trifocal tensor. International Journal of Computer Vision, 22, 125–140. CrossRefGoogle Scholar
  11. Hartley, R. (1998). Computation of the trifocal tensor. In ECCV (pp. 20–35). Google Scholar
  12. Hartley, R. I., & Zisserman, A. (2000). Multiple view geometry in computer vision. Cambridge: Cambridge University Press, ISBN:0521623049. MATHGoogle Scholar
  13. Heyden, A. (1995). Geometry and algebra of multiple projective transformations. PhD thesis, Lund University. Google Scholar
  14. Horn, B. K. P. (1987). Closed-form solution of absolute orientation using unit quaternions. Journal of the Optical Society of America, 4, 629–642. Google Scholar
  15. Kukelova, Z., Bujnak, M., & Pajdla, T. (2008). Automatic generator of minimal problem solvers. In ECCV (pp. 302–315). Google Scholar
  16. Kukelova, Z., Bujnak, M., & Pajdla, T. (2008). Polynomial eigenvalue solutions to the 5-pt and 6-pt relative pose problems. In BMVC. Google Scholar
  17. Li, H., & Hartley, R. (2006). Five-point motion estimation made easy. In ICPR (Vol. 2) Google Scholar
  18. Liu, Y., & Huang, T. (1988). A linear algorithm for motion estimation using straight line correspondences. In ICPR (pp. 213–219). Google Scholar
  19. Lowe, D. (1999). Object recognition from local scale-invariant features. In Proceedings of the international conference on computer vision (pp. 1150–1157). CrossRefGoogle Scholar
  20. Lucas, B. D., & Kanade, T. (1981). An iterative image registration technique with an application to stereo vision. In International joint conferences on oratorical intelligence (IJCAI) (pp. 1151–1156). Google Scholar
  21. Neira, J., Tardos, J. D., Horn, J., & Schmidt, G. (1999). Fusing range and intensity images for mobile robot localization. IEEE Transactions on Robotics and Automation, 15, 76–84. CrossRefGoogle Scholar
  22. Nister, D. (2004). An efficient solution to the five-point relative pose problem. IEEE Transactions on Pattern Analysis and Machine Intelligence, 26(6), 756–770. CrossRefGoogle Scholar
  23. Nister, D., Naroditsky, O., & Bergen, J. (2004). Visual odometry. In IEEE computer society conference on computer vision and pattern recognition (CVPR) (Vol. 1, pp. 652–659). Google Scholar
  24. Oliensis, J., & Werman, M. (2000). Structure from motion using points, lines, and intensities. In IEEE computer society conference on computer vision and pattern recognition (CVPR) (Vol. 2). Google Scholar
  25. Pollefeys, M., Nister, D., & et al. (2007). Detailed real-time urban 3d reconstruction from video. In IJCV. Google Scholar
  26. Pradeep, V., & Lim, J. (2010). Egomotion using assorted features. In IEEE computer society conference on computer vision and pattern recognition (CVPR) (pp. 1514–1521). Google Scholar
  27. Rosten, E., & Drummond, T. (2005). Fusing points and lines for high performance tracking. In ICCV (Vol. 2, pp. 1508–1515). Google Scholar
  28. Seitz, S., & Anandan, P. (1999). Implicit representation and scene reconstruction from probability density functions. In IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR) (Vol. 2). Google Scholar
  29. Shashua, A., & Wolf, L. (2000). On the structure and properties of the quadrifocal tensor. In Lecture notes in computer science (pp. 710–724). Google Scholar
  30. Stewénius, H., Engels, C., & Nister, D. (2006). Recent developments on direct relative orientation. Journal of Photogrammetry and Remote Sensing, 60, 284–294. CrossRefGoogle Scholar
  31. Torr, P. H. S., & Zisserman, A. (1997). Robust parameterization and computation of the trifocal tensor. Image and Vision Computing, 15, 591–605. CrossRefGoogle Scholar
  32. Triggs, B. (1999). Camera pose and calibration from 4 or 5 known 3d points. In ICCV (pp. 278–284). Google Scholar
  33. von Gioi, R. G., Jakubowicz, J., Morel, J.-M., & Randall, G. (2010). Lsd: a fast line segment detector with a false detection control. IEEE Transactions on Pattern Analysis and Machine Intelligence, 32, 722–732. CrossRefGoogle Scholar
  34. Zhang, Z. (1998). Determining the epipolar geometry and its uncertainty: a review. International Journal of Computer Vision, 27, 161–195. CrossRefGoogle Scholar
  35. Zhu, Z., Oskiper, T., Samarasekera, S., Kumar, R., & Sawhney, H. S. (2007). Ten-fold improvement in visual odometry using landmark matching. In International conference on computer vision (ICCV) (pp. 1–8). Google Scholar

Copyright information

© Springer Science+Business Media, LLC 2011

Authors and Affiliations

  1. 1.Applied Sciences GroupMicrosoft CorporationRedmondUSA
  2. 2.Honda Research InstituteMountain ViewUSA

Personalised recommendations