Advertisement

Unsupervised Crowd Counting

  • Nada ElassalEmail author
  • James H. Elder
Conference paper
  • 2k Downloads
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10115)

Abstract

Most crowd counting methods rely on training with labeled data to learn a mapping between image features and the number of people in the scene. However, the nature of this mapping may change as a function of the scene, camera parameters, illumination etc., limiting the ability of such supervised systems to generalize to novel conditions. Here we propose an alternative, unsupervised strategy. The approach is anchored on a 3D simulation that automatically learns how groups of people appear in the image. Central to the simulation is an auto-scaling step that uses the video data to infer the distribution of projected sizes of individual people detected in the scene, allowing the simulation to adapt to the signal processing parameters of the current viewing scenario. Since the simulation need only run periodically, the method is efficient and scalable to large crowds. We evaluate the method on two datasets and show that it performs well relative to supervised methods.

Keywords

Ground Plane Image Segment Perspective Projection World Coordinate System Dense Crowd 
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.

Notes

Acknowledgement

This research was supported by an NSERC Discovery research grant and by the NSERC CREATE training program in Vision Science and Applications.

Supplementary material

Supplementary material 1 (mp4 12786 KB)

References

  1. 1.
    Paragios, N., Ramesh, V.: A MRF-based approach for real-time subway monitoring. In: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, vol. 1, pp. 1034–1040 (2001)Google Scholar
  2. 2.
    Rahmalan, H., Nixon, M.S., Carter, J.N.: On crowd density estimation for surveillance. In: The Institution of Engineering and Technology Conference on Crime and Security, pp. 540–545. IET (2006)Google Scholar
  3. 3.
    Ma, R., Li, L., Huang, W., Tian, Q.: On pixel count based crowd density estimation for visual surveillance. In: 2004 IEEE Conference on Cybernetics and Intelligent Systems, vol. 1, pp. 170–173. IEEE (2004)Google Scholar
  4. 4.
    Lin, S., Chen, J., Chao, H.: Estimation of number of people in crowded scenes using perspective transformation. IEEE Trans. Syst. Man Cybern. - Part A: Syst. Hum. 31, 645–654 (2001)CrossRefGoogle Scholar
  5. 5.
    Rittscher, J., Tu, P.H., Krahnstoever, N.: Simultaneous estimation of segmentation and shape. In: IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005, vol. 2, pp. 486–493. IEEE (2005)Google Scholar
  6. 6.
    Celik, H., Hanjalic, A., Hendriks, E.A.: Towards a robust solution to people counting. In: 2006 IEEE International Conference on Image Processing, pp. 2401–2404. IEEE (2006)Google Scholar
  7. 7.
    Conte, D., Foggia, P., Percannella, G., Tufano, F., Vento, M.: A method for counting people in crowded scenes. In: 2010 Seventh IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS), pp. 225–232. IEEE (2010)Google Scholar
  8. 8.
    Conte, D., Foggia, P., Percannella, G., Vento, M.: A method based on the indirect approach for counting people in crowded scenes. In: 2010 7th IEEE International Conference on Advanced Video and Signal Based Surveillance (2010)Google Scholar
  9. 9.
    Fradi, H., Dugelay, J.L.: Low level crowd analysis using frame-wise normalized feature for people counting. In: 2012 IEEE International Workshop on Information Forensics and Security (WIFS), pp. 246–251. IEEE (2012)Google Scholar
  10. 10.
    Ryan, D., Denman, S., Sridharan, S., Fookes, C.: Scene invariant crowd counting. In: International Conference on Digital Image Computing: Techniques and Applications, pp. 237–242 (2011)Google Scholar
  11. 11.
    Zhao, T., Nevatia, R.: Bayesian human segmentation in crowded situations. In: Proceedings of the 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, vol. 2, p. II-459. IEEE (2003)Google Scholar
  12. 12.
    Kilambi, P., Masoud, O., Papanikolopoulos, N.: Crowd analysis at mass transit sites. In: IEEE Intelligent Transportation Systems Conference, ITSC 2006, pp. 753–758. IEEE (2006)Google Scholar
  13. 13.
    Kilambi, P., Ribnick, E., Joshi, A.J., Masoud, O., Papanikolopoulos, N.: Estimating pedestrian counts in groups. Comput. Vis. Image Underst. 110, 43–59 (2008)CrossRefGoogle Scholar
  14. 14.
    Fehr, D., Sivalingam, R., Morellas, V., Papanikolopoulos, N., Lotfallah, O., Park, Y.: Counting people in groups. In: Sixth IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2009, pp. 152–157. IEEE (2009)Google Scholar
  15. 15.
    Dong, L., Parameswaran, V., Ramesh, V., Zoghlami, I.: Fast crowd segmentation using shape indexing. In: IEEE 11th International Conference on Computer Vision, 2007, ICCV 2007, pp. 1–8. IEEE (2007)Google Scholar
  16. 16.
    Marana, A.N., da Fontoura Costa, L., Lotufo, R., Velastin, S.A.: Estimating crowd density with Minkowski fractal dimension. In: Proceedings of the 1999 IEEE International Conference on Acoustics, Speech, and Signal Processing, vol. 6, pp. 3521–3524. IEEE (1999)Google Scholar
  17. 17.
    Cho, S.Y., Chow, T.W.: A fast neural learning vision system for crowd estimation at underground stations platform. Neural Process. Lett. 10, 111–120 (1999)CrossRefGoogle Scholar
  18. 18.
    Cho, S.Y., Chow, T.W., Leung, C.T.: A neural-based crowd estimation by hybrid global learning algorithm. IEEE Trans. Syst. Man Cybern. Part B: Cybern. 29, 535–541 (1999)CrossRefGoogle Scholar
  19. 19.
    Huang, D., Chow, T.W., Chau, W.: Neural network based system for counting people. In: IEEE 2002 28th Annual Conference of the Industrial Electronics Society, IECON 2002, vol. 3, pp. 2197–2201. IEEE (2002)Google Scholar
  20. 20.
    Leibe, B., Seemann, E., Schiele, B.: Pedestrian detection in crowded scenes. In: IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005, vol. 1, pp. 878–885. IEEE (2005)Google Scholar
  21. 21.
    Kong, D., Gray, D., Tao, H.: A viewpoint invariant approach for crowd counting. In: 18th International Conference on Pattern Recognition, ICPR 2006, vol. 3, pp. 1187–1190. IEEE (2006)Google Scholar
  22. 22.
    Rabaud, V., Belongie, S.: Counting crowded moving objects. In: 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, vol. 1, pp. 705–711. IEEE (2006)Google Scholar
  23. 23.
    Chan, A.B., Liang, Z.S.J., Vasconcelos, N.: Privacy preserving crowd monitoring: counting people without people models or tracking. In: IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2008, pp. 1–7. IEEE (2008)Google Scholar
  24. 24.
    Jones, M.J., Snow, D.: Pedestrian detection using boosted features over many frames. In: 19th International Conference on Pattern Recognition, ICPR 2008, pp. 1–4. IEEE (2008)Google Scholar
  25. 25.
    Albiol, A., Silla, M.J., Albiol, A., Mossi, J.M.: Video analysis using corner motion statistics. In: Proceedings of the IEEE International Workshop on Performance Evaluation of Tracking and Surveillance, pp. 31–38 (2009)Google Scholar
  26. 26.
    Chan, A.B., Morrow, M., Vasconcelos, N.: Analysis of crowded scenes using holistic properties. In: Performance Evaluation of Tracking and Surveillance Workshop at CVPR, pp. 101–108 (2009)Google Scholar
  27. 27.
    Subburaman, V.B., Descamps, A., Carincotte, C.: Counting people in the crowd using a generic head detector. In: 2012 IEEE Ninth International Conference on Advanced Video and Signal-Based Surveillance (AVSS), pp. 470–475. IEEE (2012)Google Scholar
  28. 28.
    Jeong, C.Y., Choi, S., Han, S.W.: A method for counting moving and stationary people by interest point classification. In: 2013 20th IEEE International Conference on Image Processing (ICIP), pp. 4545–4548. IEEE (2013)Google Scholar
  29. 29.
    Riachi, S., Karam, W., Greige, H.: An improved real-time method for counting people in crowded scenes based on a statistical approach. In: 2014 11th International Conference on Informatics in Control, Automation and Robotics (ICINCO), vol. 2, pp. 203–212. IEEE (2014)Google Scholar
  30. 30.
    Masoud, O., Papanikolopoulos, N.P.: A novel method for tracking and counting pedestrians in real-time using a single camera. IEEE Trans. Veh. Technol. 50, 1267–1278 (2001)CrossRefGoogle Scholar
  31. 31.
    Prince, S.: Computer Vision: Models, Learning and Inference. Cambridge University Press, Cambridge (2012)CrossRefzbMATHGoogle Scholar
  32. 32.
    Ryan, D., Denman, S., Sridharan, S., Fookes, C.: An evaluation of crowd counting methods, features and regression models. Comput. Vis. Image Underst. 130, 1–17 (2015)CrossRefGoogle Scholar
  33. 33.
    Wildenauer, H., Hanbury, A.: Robust camera self-calibration from monocular images of Manhattan worlds. In: 2012 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 2831–2838. IEEE (2012)Google Scholar
  34. 34.
    Tal, R., Elder, J.H.: An accurate method for line detection and manhattan frame estimation. In: Park, J.-I., Kim, J. (eds.) ACCV 2012. LNCS, vol. 7729, pp. 580–593. Springer, Heidelberg (2013). doi: 10.1007/978-3-642-37484-5_47 CrossRefGoogle Scholar
  35. 35.
    Corral-Soto, E.R., Elder, J.H.: Automatic single-view calibration and rectification from parallel planar curves. In: Fleet, D., Pajdla, T., Schiele, B., Tuytelaars, T. (eds.) ECCV 2014. LNCS, vol. 8692, pp. 813–827. Springer, Heidelberg (2014). doi: 10.1007/978-3-319-10593-2_53 Google Scholar
  36. 36.
    Elder, J.H., Prince, S., Hou, Y., Sizintsev, M., Olevskiy, E.: Pre-attentive and attentive detection of humans in wide-field scenes. Int. J. Comput. Vis. 72, 47–66 (2007)CrossRefGoogle Scholar
  37. 37.
    Jarque, C., Bera, A.: Efficient tests for normality, homoscedasticity and serial independence of regression residuals. Econ. Lett. 6, 255–259 (1980)MathSciNetCrossRefGoogle Scholar
  38. 38.
    Zhang, Z.: Flexible camera calibration by viewing a plane from unknown orientations. In: Proceedings of the Seventh IEEE International Conference on Computer Vision, vol. 1, pp. 666–673 (1999)Google Scholar
  39. 39.
    Li, Y., Zhu, E., Zhu, X., Yin, J., Zhao, J.: Counting pedestrian with mixed features and extreme learning machine. Cogn. Comput. 6, 462–476 (2014)CrossRefGoogle Scholar
  40. 40.
    Rao, A.S., Gubbi, J., Marusic, S., Palaniswami, M.: Estimation of crowd density by clustering motion cues. Vis. Comput. 31, 1533–1552 (2015)CrossRefGoogle Scholar
  41. 41.
    Schilling, M., Watkins, A., Watkins, W.: Is human height bimodal? Am. Stat. 56, 223–229 (2002)MathSciNetCrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  1. 1.Centre for Vision ResearchYork UniversityTorontoCanada

Personalised recommendations