Skip to main content
Springer Nature Link
Log in

Tracklet association based multi-target tracking

  • Published:
Multimedia Tools and Applications Aims and scope Submit manuscript

Abstract

This paper proposes a novel multi-target tracking framework, where two different association strategies are utilized to obtain local and global tracking trajectories. Specifically, a scene self-adaptive model is first utilized to generate local trajectories by constructing the association between detection responses and tracking tracklets; then, a novel incremental linear discriminative appearance model is utilized to generate global trajectories by constructing the association between local trajectories; finally, a non-linear motion model is utilized to fill the vacancies between global trajectories to obtain continuous and smooth tracking trajectories. Experimental results conducted on PETS2009/2010, TUD-Stadtmitte, and Town Center video libraries demonstrate the proposed framework can achieve continuous and smooth tracking trajectories under the case of significant deformation, appearance change, similar appearance, motion direction change, and long-time occlusion.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Subscribe and save

Springer+ Basic
$34.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2

Similar content being viewed by others

References

  1. Andriyenko A, Schindler K, Roth S et al. (2012) Discrete-continuous optimization for multi-target tracking. IEE Conf Comput Vision Pattern Recognit 1926–1933

  2. Avidan S (2004) Support vector tracking. IEEE Trans Pattern Anal Mach Intell 26(8):1064–1072

    Article  Google Scholar 

  3. Avidan S (2005) Ensemble tracking. IEEE Conf Comput Vision Pattern Recognit 494–501

  4. Bae S, Yoon K (2014) Robust online multi-object tracking based on tracklet confidence and online discriminative appearance learning. IEEE Conf Comput Vision Pattern Recognit 1218–1225

  5. Benfold B, Reid I (2011) Stable multi-target tracking in real time surveillance video. IEEE Conf Comput Vision Pattern Recognit 3457–3464

  6. Boccignone G, Campadelli P (2010) Boosted tracking in video. IEEE Sign Process Lett 17(2):129–132

    Article  Google Scholar 

  7. Breitenstein D, Reichlin F, Leibe B et al. (2009) Robust tracking-by-detection using a detector confidence particle filter. IEEE Int Conf Comput Vision 1515–1522

  8. Chang X, Yang Y, Xing E et al. (2015) Complex event detection using semantic saliency and nearly-isotonic SVM. IEEE Conf Mach Learn 1348–1357

  9. Chang X, Yu Y, Yang Y et al. (2015) Searching persuasively: joint event detection and evidence justification with limited supervision. ACM Multimed 1–8

  10. Craciun P, Ortner M, Zerubia J et al. (2015) Joint detection and tracking of moving objects using spatio-temporal marked point processes. Winter Conf Applic Comput Vision 177–184

  11. Dalal N, B. Triggs (2005) Histograms of oriented gradients for human detection. IEEE Conf Comput Vision Pattern Recognit 886–893

  12. Felzenszwalb P, Girshick R, McAllester D, Ramanan D (2010) Object detection with discriminatively trained part-based models. IEEE Trans Pattern Anal Mach Intell 32(9):1627–1645

    Article  Google Scholar 

  13. Grabner H, Bischof H (2006) On-line boosting and vision. IEEE Comput Soc Conf Comput Vision Pattern Recognit 260–267

  14. Henriques J, Caserio R, Batista J et al. (2011) Globally optimal solution to multi-object tracking with measurements. IEEE Conf Comput Vision 2470–2477

  15. http://www.mis.tu-darmstadt.de/node/428: TUD-Stadmittte dataset

  16. http://www.cvg.rdg.ac.uk/PETS2009: Pets 2009 dataset

  17. Khan Z, Balch Z, Dellaert F (2005) MCMC-based particle filtering for tracking a variable number of interacting targets. IEEE Trans Pattern Anal Mach Intell 27(11):1805–1898

    Article  Google Scholar 

  18. Kratz L, Nishino K (2010) Tracking with local spatio-temporal motion patterns in extremely crowded scenes. IEEE Conf Comput Vision Pattern Recognit 693–700

  19. Kuo H, Nevatia R (2010) Multi-target tracking by on-line learned discriminative appearance models. IEEE Conf Comput Vision Pattern Recognit 685–692

  20. Li Y, Huang C, Nevatia R et al. (2009) Learning to associate: hybridboosted multi-target tracker for crowded scene. IEEE Conf Comput Vision Pattern Recognit 2953–2960

  21. Magee R (2004) Tracking multiple vehicles using foreground, background and motion models. Imag Vision Comput 22(2):143–155

    Article  MathSciNet  Google Scholar 

  22. Milan A, Roth S, Schindler K (2014) Continuous energy minimization for multitarget tracking. IEEE Trans Pattern Anal Mach Intell 36(1):58–72

    Article  Google Scholar 

  23. Rodriguez M, Laptev I, Sivic J et al. (2011) Density-aware person detection and tracking in crowds. IEEE Int Conf Comput Vision 2423–2430

  24. Song B, Jeng T, Staudt E et al. (2010) A stochatsic graph evolution framework for robust multi-target tracking. Europ Conf Comput Vision 605–619

  25. Song X, Shao X, Zhao H et al. (2010) An online approach: learning-semantic-scene by tracking and tracking-by-learning-semantic scene. IEEE Conf Comput Vision Pattern Recognit 739–746

  26. Wu B, Nevatia R (2007) Detection and tracking of multiple, partially occluded humans by bayesian combination of edgelet based part detectors. Int J Comput Vis 75(2):247–266

    Article  Google Scholar 

  27. Xing J, Ai H, Lao S et al. (2009) Multi-object tracking through occlusions by local tracklets filtering and global tracklets association with detection responses. IEEE Conf Comput Vision Pattern Recognit 1200–1207

  28. Xing J, Ai H, Liu L (2011) Multiple player tracking in sports video: a dual-mode two-way bayesian inference approach with progressive observation modeling. IEEE Trans Imag Process 20(6):1652–1667

    Article  MathSciNet  Google Scholar 

  29. Yan Y, Ricci E, Liu G, Sebe N (2015) Egocentric daily activity recognition via multitask clustering. IEEE Trans Image Process 24(10):2984–2995

    Article  MathSciNet  Google Scholar 

  30. Yan Y, Ricci E, Subramanian R, Liu G, Sebe N (2014) Multi-task linear discriminant analysis for multi-view action recognition. IEEE Trans Image Process 23(12):5599–5611

    Article  MathSciNet  Google Scholar 

  31. Yan Y, Ricci E, Subramanian R et al. (2013) No matter where you are: flexible graph-guided multi-task learning for multi-view head pose classification under target motion. IEEE Int Conf Comput Vision 1177–1184.

  32. Yan Y, Ricci E, Subramanian R et al. (2016) A multi-task learning framework for head pose estimation under target motion. IEEE Trans Pattern Anal Mach Intell

  33. Yan Y, Yang Y, Meng D, Liu G, Tong W, Hauptmann A, Sebe N (2015) Event oriented dictionary learning for complex event detection. IEEE Trans Image Process 24(6):1867–1878

    Article  MathSciNet  Google Scholar 

  34. Yang B, Nevatia R. (2012) Multi-target tracking by online learning of non-linear motion patterns and robust appearance models. IEEE Conf Comput Vision Pattern Recognit 1918–1925

  35. Yeh J, Hsu T (2009) Online selection of tracking features using AdBoost. IEEE Trans Circ Syst Video Technol 19(3):442–446

    Article  Google Scholar 

  36. Yu S, Yang Y, Hauptmann A et al. (2013) Harry Potter’s Marauder’s Map: localizing and tracking multiple persons-of-interest by N. discretization. IEEE Conf Comput Vision Pattern Recognit 3714–3720

  37. Zhang L, Li Y, Nevatia R et al. (2008) Global data association for multiobject tracking using network flows. IEEE Conf Comput Vision Pattern Recognit 1–8

Download references

Acknowledgments

This work is supported by Postdoctoral Foundation of China under No. 2014 M550297, Postdoctoral Foundation of Jiangsu Province under No. 1302087B, Graduate Education Reform Research and Practice Program of Jiangsu Province under No. JGZZ13_041 and JGLX15_055, Graduate Research and Innovation Program of Jiangsu under No. KYLX15_0854 and No. SJZZ15_0105.

Author information

Authors and Affiliations

  1. School of Automatic, Nanjing University of Posts and Telecommunications, Nanjing, 210046, China

    Songhao Zhu, Zhe Shi & Chengjian Sun

Authors
  1. Songhao Zhu
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Zhe Shi
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Chengjian Sun
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Songhao Zhu.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zhu, S., Shi, Z. & Sun, C. Tracklet association based multi-target tracking. Multimed Tools Appl 75, 9489–9506 (2016). https://doi.org/10.1007/s11042-015-3238-5

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11042-015-3238-5

Keywords