Abstract

This study address the problem of selecting the most informative images for multi-shot person re-identification approaches. Actually, clustering algorithms have been one of the most proposed solutions. The objective of this study is to propose a suitable clustering method for most discriminative images selection in person re-identification. Clustering methods aim to divide huge data amount into groups depending on their characteristics, so that further data processing can be easier and more manageable. Actually, clustering is concerned in multiple fields such as document treatments, video summery and image processing. Person re-identification, a new area of research deeply investigated by the vision community, would be an interest application of the clustering algorithms. Person re-identification can be defined as the process of finding the identity of an unknown person who has already been observed in a camera view. Recently, re-identifying people over large public cameras networks has become a crucial task of great importance to ensure public security. Person Re-identification approaches can be either single shot, using one image to model a person’s appearance, or multiple shot, using many images to identify a person. Actually, the real person re-identification framework is a multi-shot scenario. However, redundant images remain a challenging problem because execution time and memory consumption are significantly affected. In this study, an extensive comparison of clustering algorithms of state of art associated to a person re-identification framework is studied. Specifically, we evaluate the impact of clustering algorithms in re-identification rates. A standard re- identification framework is detailed and a synthesis of state of art methods comparison is conducted, using personal images and two standard datasets PRID_2011 and iLIDS-VID. Performing results are achieved in both person re-id rates (67.1% and 59.2%) and memory gain (92.7% and 97.8%) for Prid_2011 and iLIDS-VID datasets respectively.

Keywords:

Camera network, , clustering, , identity selection, , multi-shot, , person re-identification, , redundancy,

References

1. Agarwal, S., A. Awan and D. Roth, 2004. Learning to detect objects in images via a sparse, part-based representation. IEEE T. Pattern Anal., 26(11): 1475-1490.
   CrossRef    PMid:15521495    
2. Aravind, H., C., Rajgopal and K.P. Soman, 2010. A simple approach to clustering in excel. Int. J. Comput. Appl., 11(7): 19-25.
   
3. Ayedi, W., H. Snoussi and M. Abid, 2012. A fast multi-scale covariance descriptor for object re-identification. Pattern Recogn. Lett., 33(14): 1902-1907.
   CrossRef    
4. Bak, S., E. Corvee, F. Bremond and M. Thonnat, 2010. Person re-identification using spatial covariance regions of human body parts. Proceeding of the 7th IEEE International Conference on Advanced Video and Signal Based Surveillance. IEEE Press, Boston, pp: 435-440.
   CrossRef    
5. Bazzani, L., M. Cristani and V. Murino, 2013. Symmetry-driven accumulation of local features for human characterization and re-identification. Comput. Vis. Image Und., 117(2): 130-144.
   CrossRef    
6. Chen, D., Z. Yuan, G. Hua, N. Zheng and J. Wang, 2015. Similarity learning on an explicit polynomial kernel feature map for person re-identification. Proceeding of the IEEE Conference on Computer Vision and Pattern Recognition. Boston, MA, USA, pp: 1565-1573.
   CrossRef    
7. Chen, D., Z. Yuan, B. Chen and N. Zheng, 2016. Similarity learning with spatial constraints for person re-identification. Proceeding of the IEEE Conference on Computer Vision and Pattern Recognition. IEEE Press, Las Vegas, pp: 1268-1277.
   CrossRef    
8. Chiang, J.H. and P.Y. Hao, 2003. A new kernel-based fuzzy clustering approach: support vector clustering with cell growing. IEEE T. Fuzzy Syst., 11(4): 518-527.
   CrossRef    
9. Das, A., A. Chakraborty and A.K. Roy-Chowdhury, 2014. Consistent re-identification in a camera network. In: Fleet, D., T. Pajdla, B. Schiele and T. Tuytelaars (Eds.), Computer Vision – ECCV 2014. ECCV 2014. Lecture Notes in Computer Science, Springer, Cham, 8690: 330-345.
   CrossRef    
10. Derpanis, K.G., 2005. Mean Shift Clustering. Lecture Notes.
    
11. Georgescu, B., I. Shimshoni and P. Meer, 2003. Mean shift based clustering in high dimensions: A texture classification example. Proceeding of the 9th IEEE International Conference on Computer Vision (ICCV). Nice, France, 3: 456.
    CrossRef    
12. Geronimo, D., A.M. Lopez, A.D. Sappa and T. Graf, 2010. Survey of pedestrian detection for advanced driver assistance systems. IEEE T. Pattern Anal., 32(7): 1239-1258.
    CrossRef    PMid:20489227    
13. Gray, D. and H. Tao, 2008. Viewpoint invariant pedestrian recognition with an ensemble of localized features. In: Forsyth, D., P. Torr and A. Zisserman (Eds.), Computer Vision – ECCV 2008. Lecture Notes in Computer Science, Springer, Berlin, Heidelberg, 5302: 262-275.
    CrossRef    
14. Hassen, Y.H., T. Ouni, W. Ayedi and M. Jallouli, 2015. Mono-camera person tracking based on template matching and covariance descriptor. Proceeding of the IEEE International Conference on Computer Vision and Image Analysis Applications (ICCVIA). Sousse, Tunisia, pp: 1-4.
    CrossRef    
15. Heider, P., A. Pierre-Pierre, R. Li and C. Grimm, 2011. Local shape descriptors, A survey and evaluation. In: Laga, H., T. Schreck, A. Ferreira, A. Godil, I. Pratikakis and R.C. Veltkamp (Eds.), Proceedings of the 4th Eurographics Workshop on 3D Object Retrieval (3DOR). Eurographics Association, Llandudno, UK, April 10, 2011. pp: 49-56.
    
16. Hirzer, M., C. Beleznai, P.M. Roth and H. Bischof, 2011. Person Re-identification by Descriptive and Discriminative Classification. In: Heyden, A. and F. Kahl (Eds.), Image Analysis. SCIA 2011. Lecture Notes in Computer Science, Springer, Berlin, Heidelberg, 6688: 91-102.
    CrossRef    
17. Kanungo, T., D.M. Mount, N.S. Netanyahu, C.D. Piatko, R. Silverman and A.Y. Wu, 2002. An efficient k-means clustering algorithm: Analysis and implementation. IEEE T. Pattern Anal., 24(7): 881-892.
    CrossRef    
18. Karanam, S., M. Gou, Z. Wu, A. Rates-Borras, O.I. Camps and R.J. Radke, 2016. A comprehensive evaluation and benchmark for person re-identification: Features, metrics, and datasets. Comput. Vis. Pattern Recogn., pp: 1-19.
    
19. Li, Z., S. Chang, F. Liang, T.S. Huang, L. Cao and J.R. Smith, 2013. Learning locally-adaptive decision functions for person verification. Proceeding of the IEEE Conference on Computer Vision and Pattern Recognition. IEEE Press, Portland, pp: 3610-3617.
    CrossRef    
20. Liao, S., Y. Hu, X. Zhu and S.Z. Li, 2015. Person re-identification by local maximal occurrence representation and metric learning. Proceeding of the IEEE Conference on Computer Vision and Pattern Recognition. IEEE Press, Boston, pp: 2197-2206.
    CrossRef    
21. Lin, N.P., C.I. Chang, H.E. Chueh, H.J. Chen and W.H. Hao, 2008. A deflected grid-based algorithm for clustering analysis. WSEAS T. Comput., 7(4): 125-132.
    
22. Liu, X., M. Song, D. Tao, X. Zhou, C. Chen and J. Bu, 2014. Semi-supervised coupled dictionary learning for person re-identification. Proceeding of the IEEE Conference on Computer Vision and Pattern Recognition. IEEE Press, Columbus, pp: 3550-3557.
    CrossRef    
23. Liu, X., H. Wang, Y. Wu, J. Yang and M.H. Yang, 2015. An ensemble color model for human re-identification. Proceeding of the IEEE Winter Conference on Applications of Computer Vision (WACV). Waikoloa, HI, USA, pp: 868-875.
    CrossRef    
24. Park, H.S. and C.H. Jun, 2009. A simple and fast algorithm for K-medoids clustering. Expert Syst. Appl., 36(2/2): 3336-3341.
    
25. Wang, T., S. Gong, X. Zhu and S. Wang, 2014. Person re-identification by video ranking. Proceeding of the European Conference on Computer Vision. Zurich, Switzerland.
    CrossRef    
26. Wang, T., S. Gong, X. Zhu and S. Wang, 2016. Person re-identification by discriminative selection in video ranking. IEEE T. Pattern Anal., 38(12): 2501-2514.
    CrossRef    PMid:26829777    
27. Wong, M.A. and T. Lane, 1981. A kth nearest neighbour clustering procedure. In: Eddy, W.F. (Ed.), Computer Science and Statistics: Proceeding of the 13th Symposium on the Interface. Springer, NY, US, pp: 308-311.
    CrossRef    
28. Yang, L. and R. Jin, 2006. Distance metric learning: A comprehensive survey. Michigan State Univ., 2: 51.
    
29. Yang, Y., S. Liao, Z. Lei and S.Z. Li, 2016. Large scale similarity learning using similar pairs for person verification. Proceeding of the 13th AAAI Conference on Artificial Intelligence.
    
30. Yilmaz, A., O. Javed and M. Shah, 2006. Object tracking: A survey. ACM Comput. Surv., 38(4).
    CrossRef    
31. Zhao, R., W. Ouyang and X. Wang, 2013a. Unsupervised salience learning for person re-identification. Proceeding of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). IEEE Press, Portland, USA, pp: 3586-3593.
    CrossRef    
32. Zhao, R., W. Ouyang and X. Wang, 2013b. Person re-identification by salience matching. Proceeding of the IEEE International Conference on Computer Vision (ICCV). IEEE Press, Sydney, pp: 2528-2535.
    CrossRef    
33. Zhao, R., W. Ouyang and X. Wang, 2014. Learning mid-level filters for person re-identification. Proceeding of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). IEEE Press, Columbus, USA, pp: 144-151.
    CrossRef    
34. Zheng, L., Y. Yang and A.G. Hauptmann, 2015. Person re-identification: Past, present and future. J. Latex Class Files, 14(8): 1-20.
    

Competing interests

The authors have no competing interests.

Open Access Policy

This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Copyright

The authors have no competing interests.