Non-associative Higher-Order Markov Networks for Point Cloud Classification

  • Mohammad Najafi
  • Sarah Taghavi Namin
  • Mathieu Salzmann
  • Lars Petersson
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8693)


In this paper, we introduce a non-associative higher-order graphical model to tackle the problem of semantic labeling of 3D point clouds. For this task, existing higher-order models overlook the relationships between the different classes and simply encourage the nodes in the cliques to have consistent labelings. We address this issue by devising a set of non-associative context patterns that describe higher-order geometric relationships between different class labels within the cliques. To this end, we propose a method to extract informative cliques in 3D point clouds that provide more knowledge about the context of the scene. We evaluate our approach on three challenging outdoor point cloud datasets. Our experiments evidence the benefits of our non-associative higher-order Markov networks over state-of-the-art point cloud labeling techniques.


Non-associative Markov networks Higher-order graphical models 3D point clouds Semantic labeling 


  1. 1.
    Anand, A., Koppula, H., Joachims, T., Saxena, A.: Contextually guided semantic labeling and search for 3d point clouds. IJRR (2012)Google Scholar
  2. 2.
    Anguelov, D., Taskar, B., Chatalbashev, V., Koller, D., Gupta, D., Heitz, G., Ng, A.: Discriminative learning of markov random fields for segmentation of 3d scan data. In: CVPR, pp. 169–176 (2005)Google Scholar
  3. 3.
    Behley, J., Steinhage, V., Cremers, A.B.: Performance of histogram descriptors for the classification of 3d laser range data in urban environments. In: ICRA, pp. 4391–4398 (2012)Google Scholar
  4. 4.
    Bo, L., Lai, K., Ren, X., Fox, D.: Object recognition with hierarchical kernel descriptors. In: CVPR, pp. 1729–1736 (2011)Google Scholar
  5. 5.
    Chang, C.C., Lin, C.J.: LIBSVM: A library for support vector machines. ACM TIST 2, 27:1–27:27 (2011)Google Scholar
  6. 6.
    Daniel Munoz, N.V., Hebert, M.: Onboard contextual classification of 3-d point clouds with learned high-order markov random fields. In: ICRA (2009)Google Scholar
  7. 7.
    Gould, S., Baumstarck, P., Quigley, M., Ng, A., Koller, D.: Integrating visual and range data for robotic object detection. In: ECCV Workshop (2008)Google Scholar
  8. 8.
    Hu, H., Munoz, D., Bagnell, J.A., Hebert, M.: Efficient 3-d scene analysis from streaming data. In: ICRA, pp. 2297–2304 (2013)Google Scholar
  9. 9.
    Kim, B.S., Kohli, P., Savarese, S.: 3d scene understanding by voxel-crf. In: ICCV (2013)Google Scholar
  10. 10.
    Kohli, P., Kumar, M., Torr, P.: P3 and beyond: Move making algorithms for solving higher order functions. PAMI 31(9), 1645–1656 (2009)CrossRefGoogle Scholar
  11. 11.
    Kohli, P., Ladicky, L., Torr, P.: Robust higher order potentials for enforcing label consistency. IJCV 82(3) (2009)Google Scholar
  12. 12.
    Komodakis, N., Paragios, N.: Beyond pairwise energies: Efficient optimization for higher-order mrfs. In: CVPR, pp. 2985–2992 (2009)Google Scholar
  13. 13.
    Krähenbühl, P., Koltun, V.: Efficient inference in fully connected crfs with gaussian edge potentials. In: NIPS, pp. 109–117 (2011)Google Scholar
  14. 14.
    Ladicky, L., Russell, C., Kohli, P., Torr, P.H.S.: Inference methods for crfs with co-occurrence statistics. IJCV 103(2), 213–225 (2013)CrossRefzbMATHMathSciNetGoogle Scholar
  15. 15.
    Lafferty, J.D., McCallum, A., Pereira, F.C.N.: Conditional random fields: Probabilistic models for segmenting and labeling sequence data. In: ICML, pp. 282–289 (2001)Google Scholar
  16. 16.
    Lai, K., Bo, L., Ren, X., Fox, D.: Sparse distance learning for object recognition combining rgb and depth information. In: ICRA, pp. 4007–4013 (2011)Google Scholar
  17. 17.
    Lai, K., Fox, D.: Object recognition in 3d point clouds using web data and domain adaptation. IJRR 29(8), 1019–1037 (2010)Google Scholar
  18. 18.
    Lim, E.H., Suter, D.: 3d terrestrial lidar classifications with super-voxels and multi-scale conditional random fields. CAD 41(10), 701–710 (2009)Google Scholar
  19. 19.
    Lin, D., Fidler, S., Urtasun, R.: Holistic scene understanding for 3d object detection with rgbd cameras. In: ICCV (2013)Google Scholar
  20. 20.
    Lin, H.T., Lin, C.J., Weng, R.C.: A note on platt’s probabilistic outputs for support vector machines. ML 68(3), 267–276 (2007)Google Scholar
  21. 21.
    Lu, Y., Rasmussen, C.: Simplified markov random fields for efficient semantic labeling of 3d point clouds. In: IROS, pp. 2690–2697 (2012)Google Scholar
  22. 22.
    Munoz, D., Bagnell, J.A., Vandapel, N., Hebert, M.: Contextual classification with functional max-margin markov networks. In: CVPR, pp. 975–982 (2009)Google Scholar
  23. 23.
    Munoz, D., Bagnell, J.A., Hebert, M.: Co-inference for multi-modal scene analysis. In: Fitzgibbon, A., Lazebnik, S., Perona, P., Sato, Y., Schmid, C. (eds.) ECCV 2012, Part VI. LNCS, vol. 7577, pp. 668–681. Springer, Heidelberg (2012)CrossRefGoogle Scholar
  24. 24.
    Murphy, K.P., Weiss, Y., Jordan, M.I.: Loopy belief propagation for approximate inference: An empirical study. In: UAI, pp. 467–475 (1999)Google Scholar
  25. 25.
    Niemeyer, J., Rottensteiner, F., Soergel, U.: Contextual classification of lidar data and building object detection in urban areas. ISPRS JPRS 87, 152–165 (2014)CrossRefGoogle Scholar
  26. 26.
    Rusu, R.B., Blodow, N., Beetz, M.: Fast point feature histograms (fpfh) for 3d registration. In: ICRA, pp. 1848–1853 (2009)Google Scholar
  27. 27.
    Rusu, R.B., Cousins, S.: 3D is here: Point Cloud Library (PCL). In: ICRA (2011)Google Scholar
  28. 28.
    Shapovalov, R., Velizhev, A., Barinova, O.: Non-associative markov networks for 3d point cloud classification. In: PCV, vol. 38, pp. 103–108 (2010)Google Scholar
  29. 29.
    Shapovalov, R., Vetrov, D., Kohli, P.: Spatial inference machines. In: CVPR, pp. 2985–2992 (2013)Google Scholar
  30. 30.
    Vineet, V., Warrell, J., Torr, P.H.S.: Filter-based mean-field inference for random fields with higher-order terms and product label-spaces. In: Fitzgibbon, A., Lazebnik, S., Perona, P., Sato, Y., Schmid, C. (eds.) ECCV 2012, Part V. LNCS, vol. 7576, pp. 31–44. Springer, Heidelberg (2012)CrossRefGoogle Scholar
  31. 31.
    Wegner, J.D., Montoya-Zegarra, J.A., Schindler, K.: A higher-order crf model for road network extraction. In: CVPR, pp. 1698–1705 (2013)Google Scholar
  32. 32.
    Xiong, X., Huber, D.: Using context to create semantic 3d models of indoor environments. In: BMVC. pp. 45.1–45.11 (2010)Google Scholar
  33. 33.
    Xiong, X., Munoz, D., Bagnell, J.A.D., Hebert, M.: 3-d scene analysis via sequenced predictions over points and regions. In: ICRA (2011)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  • Mohammad Najafi
    • 1
  • Sarah Taghavi Namin
    • 1
  • Mathieu Salzmann
    • 1
  • Lars Petersson
    • 1
  1. 1.NICTAAustralian National University (ANU)CanberraAustralia

Personalised recommendations