A Method of Pedestrian Re-identification Based on Multiple Saliency Features

  • Cailing Wang
  • Yechao Xu
  • Guangwei Gao
  • Song Tang
  • Xiaoyuan Jing
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10639)

Abstract

In the field of pedestrian re-identification, the appearance of pedestrians usually varies greatly in different cameras as the pedestrians in the sequences of surveillance video may be affected by the changes of the visual angles, the postures and the light. According to this, a method of pedestrian re-identification based on multiple salient features is proposed. The traditional method of pedestrian re-identification based on salient features characterizes the weight through the difference between different samples. However, the calculation result of the traditional way is not stable enough and may vary with the rich diversity of samples. Therefore, a method based on the cellular automata is used to calculate the inherently salient features of pedestrian images. In order to make full use of the advantages of the above methods, we introduce the multi-layer cellular automata to fuse them to achieve better results in the experiment. The experimental results show that the proposed algorithm has better performance on CAVIAR4REID and iLIDS databases than the existing algorithms.

Keywords

Pedestrian re-identification Fusion of multiple salient features Multi-layer cellular automata 

Notes

Acknowledgements

The authors would like to thank the providers for their open pedestrian datasets. This work is supported by Scientific Research Foundation of China (61402237).

References

  1. 1.
    Ba̧k, S., Corvee, E., Brémond, F., Thonnat, M.: Person re-identification using spatial covariance regions of human body parts. In: Proceedings of the 7th IEEE International Conference on Advanced Video and Signal Based Surveillance, pp. 435–440. IEEE, Washington DC (2010)Google Scholar
  2. 2.
    Schwartz, W.R., Davis, L.S.: Learning discriminative appearance-based models using partial least squares. In: Proceedings of the 2009 XXII Brazilian Symposium on Computer Graphics and Image Processing, pp. 322–329. IEEE, Washington DC (2009)Google Scholar
  3. 3.
    Liu, C., Gong, S., Chen, C.L.: On-the-fly feature importance mining for person re-identification. Pattern Recogn. 47(4), 1602–1615 (2014)CrossRefGoogle Scholar
  4. 4.
    Zhao, R., Ouyang, W., Wang, X.: Unsupervised salience learning for person re-identification. In: 2013 IEEE Conference on Computer Vision and Pattern Recognition, pp. 3586–3593. IEEE, Washington DC (2013)Google Scholar
  5. 5.
    Gray, D., Tao, H.: Viewpoint invariant pedestrian recognition with an ensemble of localized features. In: Forsyth, D., Torr, P., Zisserman, A. (eds.) ECCV 2008. LNCS, vol. 5302, pp. 262–275. Springer, Heidelberg (2008). doi: 10.1007/978-3-540-88682-2_21 CrossRefGoogle Scholar
  6. 6.
    Avraham, T., Gurvich, I., Lindenbaum, M., Markovitch, S.: Learning implicit transfer for person re-identification. In: Fusiello, A., Murino, V., Cucchiara, R. (eds.) ECCV 2012. LNCS, vol. 7583, pp. 381–390. Springer, Heidelberg (2012). doi: 10.1007/978-3-642-33863-2_38 CrossRefGoogle Scholar
  7. 7.
    Qin, Y., Lu, H., Xu, Y., Wang, H.: Saliency detection via cellular automata. In: IEEE Conference on Computer Vision and Pattern Recognition, pp. 110–119. IEEE, Boston (2015)Google Scholar
  8. 8.
    Ohtsu, N.: A threshold selection method from gray-level histograms. IEEE Trans. Syst. Man Cybern. 9(1), 62–66 (1979)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  • Cailing Wang
    • 1
  • Yechao Xu
    • 1
  • Guangwei Gao
    • 1
  • Song Tang
    • 1
  • Xiaoyuan Jing
    • 1
  1. 1.Nanjing University of Posts and TelecommunicationsNanjingChina

Personalised recommendations