Abstract
High-density crowd segmentation is one of the most important components in a wide range of applications in group analysis, but the complexity and variability of the high-density crowd environment make high-density crowd behaviors segmentation facing great challenges. In this paper, we introduce a holistic approach to perform segmentation using the stability analysis based on dynamical systems. This method assumes that a grid of particles is placed over the image and moved with the underlying flow field, then computes the streaklines through those moving particles. After through a novel similarity calculation method to compute the similarity of streaklines, we capture different crowd behaviors to realize crowd segmentation. Experimental results prove the effectiveness of our algorithm.
Similar content being viewed by others
References
Mehran, R., Oyama, A., Shah, M.: Abnormal crowd behavior detection using social force model. Computer Vision and Pattern Recognition, 2009. CVPR 2009. IEEE Conference on. IEEE 935–942 (2009).
Zhang, L., Deng, T., Song, Y.: Density Estimation of Unidirectional Crowds. 2014 International Conference on Computer, Communications and Information Technology (CCIT 2014). Atlantis Press, (2014)
Mehran, R., Moore, B.E., Shah, M.: A streakline representation of flow in crowded. Scenes computer vision—ECCV. Springer, Berlin, pp. 439–452 (2010)
Alnabulsi, H., Drury, J.: Social identification moderates the effect of crowd density on safety at the Hajj. Proc. Natl. Acad. Sci. 111(25), 9091–9096 (2014)
Zhang, L., Song, M., Liu, Z.: Probabilistic graphlet cut: Exploiting spatial structure cue for weakly supervised image segmentation. Computer Vision and Pattern Recognition—CVPR, 1908–1915 (2013)
Zhang, L., Gao, Y., Xia, Y., et al.: Representative discovery of structure cues for weakly-supervised image segmentation. Multimedia 16(2), 470–479 (2014)
Zhang, L., Han, Y., Yang, Y., et al.: Discovering discriminative graphlets for aerial image categories recognition. Image Process 22(12), 5071–5084 (2013)
Zhang, L., Gao, Y., Hong, C., et al.: Feature correlation hypergraph: exploiting high-order potentials for multimodal recognition. Cybernetics 44(8), 1408–1419 (2014)
Zhang, L., Xia, Y., Ji, R., et al.: Spatial-aware object-level saliency prediction by learning graphlet hierarchies. Ind Electron 62(2), 1301–1308 (2015)
Zhang, L., Yang, Y., Gao, Y., et al.: A probabilistic associative model for segmenting weakly supervised images. Image Process 23(9), 4150–4159 (2014)
Johnson, N., Hogg, D.: Learning the distribution of object trajectories for event recognition. Image Vision Comput 14(8), 609–615 (1996)
Marques, J.S., Jorge, P.M., Abrantes, A.J.: Tracking groups of pedestrians in video sequences. Computer Vision and Pattern Recognition Workshop, CVPRW’03. Conference on. IEEE, 9:101–101 (2003)
Pellegrini, S., Ess, A., Schindler, K.: You’ll never walk alone: modeling social behavior for multi-target tracking. Computer Vision, 2009 IEEE 12th International Conference on. IEEE, pp. 261–268 (2009)
Ali, S., Shah, M.: A Lagrangian Particle Dynamics Approach for Crowd Flow Segmentation and Stability Analysis. Computer Vision and Pattern Recognition, CVPR. IEEE Conference on. IEEE, pp. 1–6 (2007)
Wijk, V., Jarke, J.: Image based flow visualization. In: SIGGRAPH (2002)
Helman, J., Hesselink, L.: Visualizing vector field topology in fluid flows. IEEE Comput Graph Appl 11, 36–46 (1991)
Landau, L., Lifshitz, E.: Advanced mechanics of fluids (1959)
Lekien, F., Marsden, J.: Tricubic interpolation in three dimensions. Int J Numer Methods Eng 63, 455–471 (2005)
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by T. Plagemann.
Rights and permissions
About this article
Cite this article
Zhang, D., Xu, J., Sun, M. et al. High-density crowd behaviors segmentation based on dynamical systems. Multimedia Systems 23, 599–606 (2017). https://doi.org/10.1007/s00530-016-0520-y
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00530-016-0520-y