Exploiting Global Self Similarity for Head-Shoulder Detection

Authors: Lae-Jeong Park, Jung-Ho Moon

Abstract:

People detection from images has a variety of applications such as video surveillance and driver assistance system, but is still a challenging task and more difficult in crowded environments such as shopping malls in which occlusion of lower parts of human body often occurs. Lack of the full-body information requires more effective features than common features such as HOG. In this paper, new features are introduced that exploits global self-symmetry (GSS) characteristic in head-shoulder patterns. The features encode the similarity or difference of color histograms and oriented gradient histograms between two vertically symmetric blocks. The domain-specific features are rapid to compute from the integral images in Viola-Jones cascade-of-rejecters framework. The proposed features are evaluated with our own head-shoulder dataset that, in part, consists of a well-known INRIA pedestrian dataset. Experimental results show that the GSS features are effective in reduction of false alarmsmarginally and the gradient GSS features are preferred more often than the color GSS ones in the feature selection.

Keywords: Pedestrian detection, cascade of rejecters, feature extraction, self-symmetry, HOG.

Digital Object Identifier (DOI): doi.org/10.5281/zenodo.1083895

References:

[1] M. Enzweiler and D. M. Gavrila, "Monocular pedestrian detection:Survey and experiments," IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 31, pp. 2179–2195,2009.
[2] D. Geronimo, A. M. Lopez, A. D. Sappa, and T. Graf, "Survey on pedestrian detection for advanced driver assistance systems," IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 32, pp. 1239–1258, 2010.
[3] K-Y K. Wong, R. H. Y. Chung, Y. L. Francis, and K-P P. Chow, "Real-time multiple head shape detection and tracking system with decentralized trackers," in Proc. Intelligent Systems Design and Application, 2006, pp. 384–389.
[4] B. Wu and R. Nevatia, "Detection and tracking of multiple, partially occluded humans by Bayesian combination of edgelet part detectors," International Journal of Computer Visions, vol. 75, pp. 247–266, 2007.
[5] M. Li, Z. Zhang, K. Huang, and T. Tan, "Estimating the number of people in crowded scenes by MID based foreground segmentation and head-shoulder detection," inProc. Int. Conf. on Pattern Recognition, 2008. pp. 1–4.
[6] X. Ding, H. Xu, P. Cui, L. Sun, and S. Yang, "A cascade SVM approach for head-shoulder detection using histograms of oriented gradients,"in Proc. IEEE Int. Symposium on Circuits and Systems, 2009, pp. 1791–1794.
[7] C. Zeng and H. Ma, "Robust head-shoulder detection by PCA-based multilevel HOG-LBP detector for people counting,"in Proc. Conf. on Pattern Recognition, 2010, pp. 2069–2072.
[8] N. Dalal and B. Triggs,"Histograms of oriented gradients for humandetection,"inProc. IEEE Conf. on Computer Vision and Pattern Recognition, 2005, pp. 886–893.
[9] P. Viola and M. J. Jones, "Rapid objection detection using a boosted cascade of simple features," inProc. IEEE Conf. on Computer Vision and Pattern Recognition, 2001, pp. 511–518.
[10] Q. Zhu, M.-C. Yeh, K.-T. Cheng, and S. Avidan, "Fast human detection using a cascade of histograms of oriented gradients," inProc. IEEE Conf. on Computer Vision and Pattern Recognition, 2006, pp. 1491–1498.
[11] P. Dollar, Z.Tu, P. Perona, and S. Belongie, "Integral channel features,"inProc. British Machine Vision Conference, 2009.
[12] S. Walk, N. Majer, K. Schindler, and B. Schiele, "New features and insights for pedestrian detection," in Proc. IEEE Conf. on Computer Vision and Pattern Recognition, 2010, pp. 1030–1037.
[13] C-H Chuang, S-S Huang, L-C Fu, and P-Y Hsiao, “Monocular multi-human detection using augmented histograms of oriented gradients,”inProc. Int. Conf. on Pattern Recognition, 2008. pp. 1–4.
[14] S. E. Fahlman and C. Lebiere, "The cascade-correlation learning architecture," in Advances in Neural Information Processing Systems 2, Margna-Kaufmann, 1990, pp. 524–532.
[15] C. Huang, H. Ai, B. Wu, and S. Lao, "Boosting nested cascade detector for multi-view face detection," in Proc. Int. Conf. on Pattern Recognition, 2004. pp. 415–418.