Multiple Instance Boost Using Graph Embedding Based Decision Stump for Pedestrian Detection

  • Junbiao Pang
  • Qingming Huang
  • Shuqiang Jiang
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5305)


Pedestrian detection in still image should handle the large appearance and stance variations arising from the articulated structure, various clothing of human as well as viewpoints. In this paper, we address this problem from a view which utilizes multiple instances to represent the variations in multiple instance learning (MIL) framework. Specifically, logistic multiple instance boost (LMIBoost) is advocated to learn the pedestrian appearance model. To efficiently use the histogram feature, we propose the graph embedding based decision stump for the data with non-Gaussian distribution. First the topology structure of the examples are carefully designed to keep between-class far and within-class close. Second, K-means algorithm is adopted to fast locate the multiple decision planes for the weak classifier. Experiments show the improved accuracy of the proposed approach in comparison with existing pedestrian detection methods, on two public test sets: INRIA and VOC2006’s person detection subtask [1].


Support Vector Machine Multiple Instance Linear Support Vector Machine Histogram Feature Pedestrian Detection 
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.


  1. 1.
    Everingham, M., Zisserman, A., Williams, C.K.I., Gool, L.V.: The PASCAL Visual Object Classes Challenge (VOC 2006) Results (2006),
  2. 2.
    Gavrila, D.M.: Pedestrian detection from a moving vehicle. In: Vernon, D. (ed.) ECCV 2000. LNCS, vol. 1843, pp. 37–49. Springer, Heidelberg (2000)CrossRefGoogle Scholar
  3. 3.
    Bissacco, A., Yang, M., Soatto, S.: Detection human via their pose. In: Proc. NIPS (2006)Google Scholar
  4. 4.
    Dalal, N., Triggs, B.: Histograms of oriented gradients for human detection. In: Proc. CVPR, vol. I, pp. 886–893. IEEE, Los Alamitos (2005)Google Scholar
  5. 5.
    Papageorgiou, P., Poggio, T.: A trainable system for object detection. IJCV, 15–33 (2000)Google Scholar
  6. 6.
    Viola, P., Jones, M., Snow, D.: Detecing pedestrians using patterns of motion and appearance. In: Proc. ICCV (2003)Google Scholar
  7. 7.
    Maron, O., Lozanno-Perez, T.: A framework for multiple-instance learning. In: Proc. NIPS, pp. 570–576 (1998)Google Scholar
  8. 8.
    Dietterich, T., Lathrop, R., Lozano-Perez, T.: Solving the multiple instance problem with axis-parallel rectangles. Artifical intelligence, 31–71 (1997)Google Scholar
  9. 9.
    Xu, X., Frank, E.: Logistic regression and boosting for labeled bags of instances. In: Dai, H., Srikant, R., Zhang, C. (eds.) PAKDD 2004. LNCS (LNAI), vol. 3056, pp. 272–281. Springer, Heidelberg (2004)Google Scholar
  10. 10.
    Sugiyama, M.: local fisher discriminat analysis for supervised dimensionality reduction. In: Proc. ICML (2006)Google Scholar
  11. 11.
    Monhan, A., Papageorgiou, C., Poggio, T.: Example-based object detection in images by componets. IEEE Trans. PAMI 23, 349–360 (2001)Google Scholar
  12. 12.
    Lowe, D.G.: Distinctive image features from scale-invariant keypoints, 91–110 (2004)Google Scholar
  13. 13.
    Zhu, Q., Avidan, S., Yeh, M.C., Cheng, K.T.: Fast human detection using a cascade of histograms of oriented gradients. In: Proc. CVPR, vol. 2, pp. 1491–1498. IEEE, Los Alamitos (2006)Google Scholar
  14. 14.
    Tuzel, O., Porikli, F., Meer, P.: Human detection via classification on riemannina manifolds. In: Proc. CVPR. IEEE, Los Alamitos (2007)Google Scholar
  15. 15.
    Zisserman, A., Schmid, C., Mikolajczyk, K.: Human detection based on a probabilistic assembly of robust part detectors. In: Pajdla, T., Matas, J(G.) (eds.) ECCV 2004. LNCS, vol. 3021, pp. 69–82. Springer, Heidelberg (2004)Google Scholar
  16. 16.
    Leibe, B., Seemann, E., Schiele, B.: Pedestrian detection in crowed scenes. In: Proc. CVPR, pp. 878–885. IEEE, Los Alamitos (2005)Google Scholar
  17. 17.
    Tran, D., Forsyth, D.A.: Configuration estimates improve pedestrian finding. In: Proc. NIPS (2007)Google Scholar
  18. 18.
    Felzenszwalb, P., Mcallester, D., Ramanan, D.: A discriminatively trained, multiscale, deformable part model. In: Proc. CVPR. IEEE, Los Alamitos (2008)Google Scholar
  19. 19.
    Maron, O., Ratan, A.: Multiple-instance learning for natural scene classification. In: Proc. ICML (1998)Google Scholar
  20. 20.
    Viola, P., Platt, J.C., Zhang, C.: Multiple instance boosting for object detection. In: Proc. NIPS (2006)Google Scholar
  21. 21.
    Porikli, F.M.: Integral histogram: a fast way to extract histogram in cartesian space. In: Proc. CVPR, pp. 829–836. IEEE, Los Alamitos (2005)Google Scholar
  22. 22.
    Liu, C., Shum, H.Y.: Kullback-leibler boosting. In: Proc. CVPR, pp. 587–594 (2003)Google Scholar
  23. 23.
    Laptev, I.: Improvements of object detection using boosted histograms. In: BMVC (2006)Google Scholar
  24. 24.
    Viola, P., Jones, M.: Rapid object detection using a boosted cascade of simple features. In: Proc. CVPR. IEEE, Los Alamitos (2001)Google Scholar
  25. 25.
    Huang, C., Ai, H., Wu, B., Lao, S.: Boosting nested cascade detector for multi-view face detection. In: Proc. ICPR. IEEE, Los Alamitos (2004)Google Scholar
  26. 26.
    Xiao, R., Zhu, H., Sun, H., Tang, X.: Dynamic cascades for face detection. In: Proc. ICCV. IEEE, Los Alamitos (2007)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2008

Authors and Affiliations

  • Junbiao Pang
    • 1
    • 2
    • 3
  • Qingming Huang
    • 1
    • 2
    • 3
  • Shuqiang Jiang
    • 2
    • 3
  1. 1.Graduate university of Chinese Academy of SciencesBeijingChina
  2. 2.Key Lab. of Intelligent Information ProcessingChinese Academy of Sciences (CAS) 
  3. 3.Institute of Computing TechnologyCASBeijingChina

Personalised recommendations