Advertisement

International Journal of Computer Vision

, Volume 97, Issue 2, pp 123–147 | Cite as

Energy-Based Geometric Multi-model Fitting

  • Hossam Isack
  • Yuri BoykovEmail author
Article

Abstract

Geometric model fitting is a typical chicken-&-egg problem: data points should be clustered based on geometric proximity to models whose unknown parameters must be estimated at the same time. Most existing methods, including generalizations of RANSAC, greedily search for models with most inliers (within a threshold) ignoring overall classification of points. We formulate geometric multi-model fitting as an optimal labeling problem with a global energy function balancing geometric errors and regularity of inlier clusters. Regularization based on spatial coherence (on some near-neighbor graph) and/or label costs is NP hard. Standard combinatorial algorithms with guaranteed approximation bounds (e.g. α-expansion) can minimize such regularization energies over a finite set of labels, but they are not directly applicable to a continuum of labels, e.g. \({\mathcal{R}}^{2}\) in line fitting. Our proposed approach (PEaRL) combines model sampling from data points as in RANSAC with iterative re-estimation of inliers and models’ parameters based on a global regularization functional. This technique efficiently explores the continuum of labels in the context of energy minimization. In practice, PEaRL converges to a good quality local minimum of the energy automatically selecting a small number of models that best explain the whole data set. Our tests demonstrate that our energy-based approach significantly improves the current state of the art in geometric model fitting currently dominated by various greedy generalizations of RANSAC.

Keywords

Geometric models α-Expansion Graph cuts Sampling Labeling Parameter estimation 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Beis, J. S., & Lowe, D. G. (1997). Shape indexing using approximate nearest-neighbour search in high-dimensional spaces. In CVPR (pp. 1000–1006). Google Scholar
  2. Birchfield, S., & Tomasi, C. (1999). Multiway cut for stereo and motion with slanted surfaces. In ICCV. Google Scholar
  3. Bishop, C. M. (2006). Pattern recognition and machine learning. Berlin: Springer. zbMATHGoogle Scholar
  4. Boult, T., & Brown, L. G. (1991). Factorization-based segmentation of motions. In IEEE workshop on visual motion. Google Scholar
  5. Boykov, Y., Veksler, O., & Zabih, R. (2001). Fast approximate energy minimization via graph cuts. In PAMI. Google Scholar
  6. Chin, T.-J., Wang, H., & Suter, D. (2009). Robust fitting of multiple structures: the statistical learning approach. In International Conference on Computer Vision (ICCV). Google Scholar
  7. Chin, T.-J., Yu, J., & Suter, D. (2010). Accelerated hypothesis generation for multi-structure robust fitting. In European Conference on Computer Vision (ECCV). Google Scholar
  8. Chum, O., Matas, J., & Kittler, J. (2003). Locally optimized RANSAC. In LNCS: Vol. 2781. Pattern recognition (pp. 236–243). CrossRefGoogle Scholar
  9. Comaniciu, D., & Meer, P. (2002). Mean shift: a robust approach toward feature space analysis. In PAMI. Google Scholar
  10. Costeira, J., & Kanade, T. (1995). A multi-body factorization method for motion analysis. In ICCV. Google Scholar
  11. Delong, A., Osokin, A., Isack, H., & Boykov, Y. (2011). Fast approximate energy minization with label costs. International Journal of Computer Vision (accepted). Earlier version is in CVPR 2010. doi: 10.1007/s11263-011-0437-z
  12. Faugeras, O., & Luong, Q.-T. (2004). The geometry of multiple images. Cambridge: MIT Press. Google Scholar
  13. Figueiredo, M. A., & Jain, A. K. (2002). Unsupervised learning of finite mixture models. IEEE Transactions on Pattern Analysis and Machine Intelligence, 24(3), 381–396. CrossRefGoogle Scholar
  14. Fischler, M. A., & Bolles, R. C. (1981). Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography. In CACM. Google Scholar
  15. Gruber, A., & Weiss, Y. (2006). Incorporating non-motion cues into 3D motion segmentation. In European Conference on Computer Vision (ECCV). Google Scholar
  16. Hartley, R. (1997). In defense of the eight-point algorithm. IEEE Transactions on Pattern Analysis and Machine Intelligence, 19(6), 580–593. CrossRefGoogle Scholar
  17. Hartley, R., & Zisserman, A. (2003). Multiple view geometry in computer vision. Cambridge: Cambridge University Press. Google Scholar
  18. Isack, H. (2009). Spatially coherent multi-model fitting. MS Thesis, CS Dept., University of Western Ontario, London, Canada. Google Scholar
  19. Leclerc, Y. G. (1989). Constructing simple stable descriptions for image partitioning. International Journal of Computer Vision, 3(1), 73–102. CrossRefGoogle Scholar
  20. Li, H. (2007). Two-view motion segmentation from linear programming relaxation. In CVPR. Google Scholar
  21. Lowe, D. G. (2004). Distinctive image features from scale-invariant keypoints. In IJCV. Google Scholar
  22. Ma, Y., Soatto, S., Kosecka, J., & Sastry, S. (2003). An invitation to 3D vision: from images to geometric models. Berlin: Springer. Google Scholar
  23. Muja, M., & Lowe, D. G. (2009). Fast approximate nearest neighbors with automatic algorithm configuration. In VISAPP. Google Scholar
  24. Olsson, C., Enqvist, O., & Kahl, F. (2008). A polynomial-time bound for matching and registration with outliers. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Anchorage, USA. Google Scholar
  25. Rother, C., Kolmogorov, V., & Blake, A. (2004). Grabcut—interactive foreground extraction using iterated graph cuts. In ACM Transactions on Graphics (SIGGRAPH), August 2004. Google Scholar
  26. Schindler, K., & Suter, D. (2006). Two-view multibody structure-and-motion with outliers through model selection. IEEE Transactions on Pattern Analysis and Machine Intelligence, 28(6), 983–995. CrossRefGoogle Scholar
  27. Toldo, R., & Fusiello, A. (2008). Robust multiple structures estimation with J-linkage. In ECCV. Google Scholar
  28. Tomasi, C., & Kanade, T. (1992). Shape and motion from image streams under orthography: a factorization method. In IJCV. Google Scholar
  29. Torr, P., & Zisserman, A. (2000). MLESAC: a new robust estimator with application to estimating image geometry. Journal of Computer Vision and Image Understanding, 78(1), 138–156. CrossRefGoogle Scholar
  30. Torr, P. H. S. (1998). Geometric motion segmentation and model selection. Philosophical Transactions of the Royal Society A, 1321–1340. Google Scholar
  31. Torr, P. H. S., & Murray, D. W. (1994). Stochastic motion clustering. In LNCS: Vol. 801. European Conference on Computer Vision (ECCV), Stockholm, Sweden (pp. 328–337). Google Scholar
  32. Tron, R., & Vidal, R. (2007). A benchmark for the comparison of 3-d motion segmentation algorithms. In CVPR. Google Scholar
  33. Vidal, R., Tron, R., & Hartley, R. (2008). Multiframe motion segmentation with missing data using powerfactorization and GPCA. In IJCV. Google Scholar
  34. Vincent, E., & Laganiere, R. (2001). Detecting planar homographies in an image pair. In ISPA, June. Google Scholar
  35. Wills, J., Agarwal, S., & Belongie, S. (2003). What went where. In CVPR03 (pp. 37–44). Google Scholar
  36. Yan, J., & Pollefeys, M. (2006). A general framework for motion segmentation: independent, articulated, rigid, non-rigid, degenerate, and non-degenerate. In European Conference on Computer Vision (ECCV). Google Scholar
  37. Zabih, R., & Kolmogorov, V. (2004). Spatially coherent clustering with graph cuts. In CVPR, June. Google Scholar
  38. Zhu, S. C., & Yuille, A. (1996). Region competition: unifying snakes, region growing, and Bayes/MDL for multiband image segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 18(9), 884–900. CrossRefGoogle Scholar
  39. Zrour, R., Kenmochi, Y., Talbot, H., Buzer, L., Hamam, Y., Shimizu, I., & Sugimoto, A. (2011). Optimal consensus set for digital line and plane fitting. International Journal of Imaging Systems and Technology, 21, 45–57. CrossRefGoogle Scholar
  40. Zuliani, M., Kenney, C., & Manjunath, B. (2005). The multiransac algorithm and its application to detect planar homographies. In ICIP. Google Scholar

Copyright information

© Springer Science+Business Media, LLC 2011

Authors and Affiliations

  1. 1.Computer Science DepartmentUniversity of Western OntarioLondonCanada

Personalised recommendations