Data-Driven and Collision-Free Hybrid Crowd Simulation Model for Real Scenario

  • Qingrong Cheng
  • Zhiping Duan
  • Xiaodong Gu
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11307)


In order to take into account evading mechanism and make more realistic simulation results, we propose a data-driven and collision-free hybrid crowd simulation model in this paper. The first part of the model is a data-driven process in which we introduce an algorithm called MS-ISODATA (Main Streams Iterative Self-organizing Data Analysis) to learn motion patterns from real scenarios. The second part introduces an agent-based collision-free mechanism in which a steering approach is improved and this part uses the output from the first part to guide its agents. The hybrid simulation model we propose can reproduce simulated crowds with motion features of real scenarios, and it also enables agents in simulation evade from mutual collisions. The simulation results show that the hybrid crowd simulation model mimics the desired crowd dynamics well. According to a collectiveness measurement, the simulation results and real scenarios are very close. Meanwhile, it reduces the number of virtual crowd collisions and makes the movement of the crowd more effective.


Crowd simulation Data-driven Main streams MS-ISODATA Collision-free Collectiveness 



This work is supported by national natural science foundation of China under grants 61771145 and 61371148.


  1. 1.
    Chan, A.B., Liang, Z.S.J., Vasconcelos, N.: Privacy preserving crowd monitoring: counting people without people models or tracking. In: 26th IEEE Conference on Computer Vision and Pattern Recognition, pp. 1–7. IEEE Press, Anchorage (2008)Google Scholar
  2. 2.
    Huang, X., Wang, W., Shen, G., Feng, X., Kong, X.: Crowd activity classification using category constrained correlated topic model. KSII Trans. Internet Inf. Syst. 10, 5530–5546 (2016)Google Scholar
  3. 3.
    Bera, A., Manocha, D.: Realtime multilevel crowd tracking using reciprocal velocity obstacles. In: 22nd IEEE International Conference on Pattern Recognition, pp. 4164–4169. IEEE Press, Stockholm (2014)Google Scholar
  4. 4.
    Cui, J., Liu, W., Xing, W.: Crowd behaviors analysis and abnormal detection based on surveillance data. J. Vis. Lang. Comput. 25(6), 628–636 (2014)CrossRefGoogle Scholar
  5. 5.
    Shao, J., Loy, C.C., Wang, X.: Learning scene-independent group descriptors for crowd understanding. IEEE Trans. Circ. Syst. Video Technol. 27(6), 1290–1303 (2017)CrossRefGoogle Scholar
  6. 6.
    Martinez-Gil, F., Lozano, M., Fernández, F.: Emergent behaviors and scalability for multi-agent reinforcement learning-based pedestrian models. Simu. Model. Pract. Theory 74, 117–133 (2017)CrossRefGoogle Scholar
  7. 7.
    Junior, J.C.S.J., Musse, S.R., Jung, C.R.: Crowd analysis using computer vision techniques. IEEE Sig. Process. Mag. 27(5), 66–77 (2010)Google Scholar
  8. 8.
    Reynolds, C.W.: Flocks, herds and schools: a distributed behavioral model. ACM SIGGRAPH Comput. Graph. 21(4), 25–34 (1987)CrossRefGoogle Scholar
  9. 9.
    Helbing, D., Molnar, P.: Social force model for pedestrian dynamics. Phys. Rev. E 51(5), 4282 (1995)CrossRefGoogle Scholar
  10. 10.
    Helbing, D., Farkas, I., Vicsek, T.: Simulating dynamical features of escape panic. Nature 407(6803), 487–490 (2000)CrossRefGoogle Scholar
  11. 11.
    Duan, Z., Gu, X.: Animal group behavioral model with evasion mechanism. In: IEEE International Joint Conference on Neural Networks, pp. 1167–1172. IEEE Press, Beijing (2014)Google Scholar
  12. 12.
    Lerner, A., Chrysanthou, Y., Lischinski, D.: Crowds by example. Comput. Graph. Forum 26(3), 655–664 (2007)CrossRefGoogle Scholar
  13. 13.
    Lee, K.H., Choi, M.G., Hong, Q., Lee, J.: Group behavior from video: a data-driven approach to crowd simulation. In: Proceedings of the 2007 ACM SIGGRAPH/Eurographics Symposium on Computer Animation, pp. 109–118. ACM Press, San Diego (2007)Google Scholar
  14. 14.
    Karamouzas, I., Skinner, B., Guy, S.J.: Universal power law governing pedestrian interactions. Phys. Rev. Lett. 113(23), 238701 (2014)CrossRefGoogle Scholar
  15. 15.
    Benfold, B., Reid, I.: Stable multi-target tracking in real-time surveillance video. In: 29th IEEE Conference on Computer Vision and Pattern Recognition, pp. 3457–3464. IEEE Press, Colorado (2011)Google Scholar
  16. 16.
    Berclaz, J., Fleuret, F., Fua, P.: Multiple object tracking using flow linear programming. In: 12th IEEE International Workshop on Performance Evaluation of Tracking and Surveillance (PETS-Winter), pp. 1–8. IEEE Press, Snowbird (2009)Google Scholar
  17. 17.
    Tomasi, C.: Detection and tracking of point features. Technical report 91(21), 9795–9802 (1991)Google Scholar
  18. 18.
    Zhou, B., Tang, X., Wang, X.: Measuring crowd collectiveness. In: 31st IEEE Conference on Computer Vision and Pattern Recognition, pp. 3049–3056. IEEE Press, Portland (2013)Google Scholar
  19. 19.
    Pellegrini, S., Ess, A., Schindler, K., Van Gool, L.: You’ll never walk alone: Modeling social behavior for multi-target tracking. In: 12th IEEE International Conference on Computer Vision, pp. 261–268. Kyoto (2009)Google Scholar
  20. 20.
    Ball, G.H., Hall, D.J.: ISODATA, a novel method of data analysis and pattern classification. Stanford Research Institute, AD-699616 (1965)Google Scholar
  21. 21.
    Ondřej, J., Pettré, J., Olivier, A.H., Donikian, S.: A synthetic-vision based steering approach for crowd simulation. ACM Trans. Graph. (TOG) 29(4), 123 (2010)CrossRefGoogle Scholar
  22. 22.
    Zhou, B., Wang, X., Tang, X.: Understanding collective crowd behaviors: learning a mixture model of dynamic pedestrian-agents. In: 30th IEEE Conference on Computer Vision and Pattern Recognition, pp. 2871–2878. IEEE Press, Providence (2012)Google Scholar
  23. 23.
    Guy, S., Van Den Berg, J., Liu, W., Lau, R., Lin, M.C.: A statistical similarity measure for aggregate crowd dynamics. ACM Trans. Graph. (TOG) 31(6), 190 (2012)CrossRefGoogle Scholar
  24. 24.
    Wong, S.M., Yao, Y.Y.: A statistical similarity measure. In: Proceedings of the 10th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, pp. 3–12. ACM Press, New Orleans (1987)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2018

Authors and Affiliations

  1. 1.Department of Electronic EngineeringFudan UniversityShanghaiChina

Personalised recommendations