Total views : 497

Fast Multi-Object Tracking-by-Detection Using Tracker Affinity Matrix


  • PAWLIN Technologies Ltd., Dubna, Russian Federation


Objectives: This study is an attempt to develop a fast real-time multi-object tracking algorithm with competitive precision and accuracy in crowded scenes. Methods/Statistical Analysis: This research extends multi-object tracking-by-detection framework by utilizing physical characteristics of trackers and their affinity for tracker guidance, in addition to detections and representation models. Guidance is done by particle filter which weights are updated through reworked observation model, featuring a term based upon tracker affinity. Tracker affinity is also used for tracker grouping that removes redundant trackers following the same target, and triggering a special propagation mode during occlusions that prevent identity switches. Findings: Our research has shown that trackers affinity matrix and algorithm features based on it yield substantial extra accuracy for tracking-by-detection framework, while taking a minor fraction of processing time. Improvements/Applications: In this paper we have introduced an approach that makes it possible to use less accurate, but faster detectors and representation model classifiers, enabling real-time processing, while keeping competitive precision and accuracy.


Multi-Object Tracking, Tracking-by-Detection, Particle Filter, Pedestrian Detection, Particle Filtering, Online Learning, Tracker Affinity, Surveillance.

Full Text:

 |  (PDF views: 297)


  • Welsh BC, Farrington DP. Public Area CCTV and Crime Prevention: An Updated Systematic Review and Meta-Analysis. Justice Quarterly. 2009 October; 26(4):716-45.
  • Wang L, Hu W, Tan T. Recent developments in human motion analysis. Pattern Recognit. 2003; 36(3):585-601.
  • La Vigne N, Lowry S, Markman J, Dwyer A. Evaluating the use of public surveillance cameras for crime control and prevention. Washington, DC: US Department of Justice, Office of Community Oriented Policing Services. Urban Institute, Justice Policy Center. 2011.
  • Burke RR. The Third Wave of Marketing Intelligence. In: Retailing in the 21st Century. Krafft M and Mantrala MK (Eds.) Springer Berlin Heidelberg, 2010; p.159-71.
  • Breitenstein MD, Reichlin F, Leibe B, Koller-Meier E, Gool LV, Online Multiperson Tracking-by-Detection from a Single, Uncalibrated Camera. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2011; 33(9):1820-33.
  • Choi W. Near-Online Multi-target Tracking with Aggregated Local Flow Descriptor. ICCV, 2015 1-11. Date accessed: 20/04/2016: Available from:
  • Grabner H, Leistner C, Bischof H. Semi-supervised on-line boosting for robust tracking. Computer Vision – ECCV, Series Lecture Notes in Computer Science. 2008; 5302:234-47.
  • Wang B, Wang G, Chan KL, Wang L. Tracklet association with online target-specific metric learning. Proc. of IEEE Conf. CVPR. 2015; p. 1234-41.
  • Leal-Taixe L, Milan A, Reid I, Roth S, Schindler K. MOT Challenge 2015: Towards a benchmark for multi-target tracking. arXiv:1504.01942 [cs], 2015 April.
  • Wu Y, Lim J, Yang MH. Online object tracking: A benchmark. Proc. Comput. Vis. Pattern Recognit. 2013; p. 2411-18.
  • Zhong W, Lu H, Yang M-H. Robust Object Tracking via Sparsity-based Collaborative Model. Proc. IEEE Conf. on CVPR. 2012; p. 1838-45. Date accessed: 23/04/2016: Available from:
  • Hare S, Saffari A, Torr PHS. Struck: Structured Output Tracking with Kernels. Proc. IEEE International Conference on Computer Vision (ICCV). 2011; p. 263-70.
  • Comaniciu D, Ramesh V, Meer P. Kernel-Based Object Tracking. PAMI. 2003; 25(5):564-77.
  • Sevilla-Lara L, Learned-Miller E. Distribution Fields for Tracking. Proc. IEEE International Conference on CVPR. 2012; p. 1910-17.
  • Jia X, Lu H, Yang M-H. Visual Tracking via Adaptive Structural Local Sparse Appearance Model. Proc. IEEE International Conference on CVPR. 2012; p. 1822-29.
  • Mei X, Ling H. Robust Visual Tracking using L1 Minimization. Proc. IEEE 12th International Conference on ICCV. 2009; p. 1436-43.
  • Kuhn H. The Hungarian method for the assignment problem. Naval Research Logistics Quarterly. 1955; 2:83-87.
  • Cormen ThH, Leiserson ChE, Rivest RL, Stein C. MIT Press: Chapter 21: Data structures for Disjoint Sets. Introduction to Algorithms (2nd ed.). 2001; p. 498-524.
  • Sheshasayee A, Lakshmi JNV. Comparison of Machine Learning Algorithm on Map Reduction for Performance Improvement in Big Data. IJST. 2015 November; 8(29).
  • Mekhtiche MA, et al. Real Time Object Detection & Tracking over a Mobile Platform. IJST. 2015 November; 8(32).
  • Ferryman JM. (Ed.) Snowbird, USA: Proceedings of the 12th IEEE International Workshop on Performance Evaluation of Tracking and Surveillance. IEEE Computer Society. 2009.
  • Bernardin K, Stiefelhagen R. Evaluating multiple object tracking performance: The CLEAR MOT metrics. J. Image and Video Processing. 2008; 3:1-10.
  • Skribtsov PV, Surikov SO, Yakovlev MA. Background Image Estimation with MRF and DBSCAN Algorithms. 2015; 8(S10):1-6.


  • There are currently no refbacks.

Creative Commons License
This work is licensed under a Creative Commons Attribution 3.0 License.