Abstract
Multiple object tracking (MOT) involves consistent labeling of objects in a given scene. A scene consists of multiple frames and within each frame rectangular subregions are specified as objects of interest. The task is to label the same object across frames with same identifier. However challenges in this setting involve, change in posture of the object, mild background change in the object region, occlusion, lighting changes, speed of movement and other such critical parameters. MOT is important because of its various applications in mobile robots, autonomous driving, and video surveillance analysis. There a number of neural network based methods which add modules based on property of interest such as a sub-network for velocity, a network for physical motion characteristics and networks based on pixel and edge information characteristics. However they have difficulty dealing with long duration occlusions as well as generalization issues due to millions of parameters and implicit overfitting. We present a new idea called, Temporal Local Slicing (TLS) that obtains local information across frames for a given subregion in the object vectorization step. The vectorization involves histogram of pixel intensities for red, blue and green channels of the sub region. We have performed a total of five experiments and observed the effectiveness of TLS and also a new idea of Gossip vectorization in Multiple object tracking. The object recognition accuracy of TLS vectors is 99.5% and mAP score of 99.1% on train and test partition of a video scene. However the MOT specific scores have been MOTA 56%, IDF1 72%, Recall 56.7%, Precision 98.5% and LOCA 91.9%. These are non-trivial scores indicating potential value in the idea of TLS vectorization.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Michael, B.D., Fabian, R., Bastian, L., Esther, K.M., Luc, V.G.: Robust tracking-by-detection using a detector confidence particle filter. In: 2009 IEEE 12th International Conference on Computer Vision, May 2010. https://doi.org/10.1109/ICCV.2009.5459278
Jerome, B., Francois, F., Pascal, F.: Robust people tracking with global trajectory optimization. In: 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR’06), June 2006. https://doi.org/10.1109/CVPR.2006.258
Jerome, B., Francois, F., Pascal, F.: Multiple object tracking using k-shortest paths optimization. IEEE Trans. Pattern Anal. Mach. Intell. 1806–1819 (2011). https://doi.org/10.1109/TPAMI.2011.21
Joshua, C., Matthew, S., Goldgof, D.B., Deborah, S.B., Rangachar, K.: Understanding transit scenes: a survey on human behavior-recognition algorithms. IEEE Trans. Intell. Transport. Syst. 206–224 (2010). https://doi.org/10.1109/TITS.2009.2030963
Patrick, D., et al.: Mot20: a benchmark for multi object tracking in crowded scenes (2020). https://arxiv.org/abs/2003.09003
Forney, G.D.: The Viterbi algorithm. Proc. IEEE 61(3), 268–278 (1973). https://doi.org/10.1109/PROC.1973.9030
Uchiyama, H., Marchand, E.: Object detection and pose tracking for augmented reality: recent approaches, November 2012. https://hal.inria.fr/hal-00751704/
Kullback, S., Leibler, R.A.: On information and sufficiency. Ann. Math. Stat. 22(1), 79 – 86 (1951). https://doi.org/10.1214/aoms/1177729694
Wenhan, L., Junliang, X., Anton, M., Xiaoqin, Z., Wei, L., Tae-Kyun, K.: Multiple object tracking: a literature review. Artif. Intell. ELSEVIER 293(103448) (2020). https://doi.org/10.1016/j.artint.2020.103448
Xin, L., Kejun, W., Wei, W., Yang, L.: A multiple object tracking method using Kalman filter. In: 2010 IEEE International Conference on Information and Automation, pp. 1862–1866 (2010). https://doi.org/10.1109/ICINFA.2010.5512258
Luiten, J., et al.: HOTA: a higher order metric for evaluating multi-object tracking. Int. J. Comput. Vision, 1–31 (2020). https://doi.org/10.1007/s11263-020-01375-2
Hilda, M.F., Adriane, E.S.: Looking at the center of the targets helps multiple object tracking. J. Vision 10 (2010). https://doi.org/10.1167/10.4.19
Morefield, C.L.: Application of 0-1 integer programming to multitarget tracking problems. IEEE Trans. Autom. Control AC-22(3), 302–312 (1977). https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1101500
Storms, P., Spieksma, F.: An LP-based algorithm for the data association problem in multitarget tracking. In: Proceedings of the Third International Conference on Information Fusion (2000). https://doi.org/10.1109/IFIC.2000.862699
Pedregosa, F., et al.: Scikit-learn: machine learning in Python. J. Mach. Learn. Res. 12, 2825–2830 (2011)
Prashanth, K., Kalidas, Y., Jay, R.B.K., Sai, A.P.K., Aakash, D.: An algorithm for semantic vectorization of video scenes - applications to retrieval and anomaly detection. In: International Conference on Computer Vision and Image Processing (CVIP) 2020, vol. 1378, pp. 369–381 (2020). https://doi.org/10.1007/978-981-16-1103-2_31
Kumar, R., Guillaume, C., Monique, T.: Multiple object tracking by efficient graph partitioning. In: Brown, M.S., Cham, T.-J., Matsushita, Y. (eds.) ACCV - 12th Asian Conference on Computer Vision, November 2014, Singapore, Singapore. ffhal-01061450f (2014). https://hal.inria.fr/hal-01061450
Nadia, R., Imas, S.S.: Determination of optimal epsilon (EPS) value on DBScan algorithm to clustering data on peatland hotspots in Sumatra. IOP Conf. Ser. Earth Environ. Sci. https://iopscience.iop.org/article/10.1088/1755-1315/31/1/012012/pdf
Redmon, J., Divvala, S., Girshick, R., Farhadi, A.: You only look once: unified, real-time object detection. In: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 779–788 (2016). https://doi.org/10.1109/CVPR.2016.91
Hu, W., Li, X., Luo, W., Zhang, X., Maybank, S., Zhang, Z.: Single and multiple object tracking using log-Euclidean Riemannian subspace and block-division appearance model. IEEE Trans. Pattern Anal. Mach. Intell. (2012). https://doi.org/10.1109/TPAMI.2012.42
Nicolai, W., Alex, B., Dietrich, P.: Simple online and realtime tracking with a deep association metric (2017). https://arxiv.org/abs/1703.07402
Xiaogang, W.: Intelligent multi-camera video surveillance: a review. Pattern Recogn. Lett. ELSEVIER, 3–19 (2012). https://doi.org/10.1016/j.patrec.2012.07.005, https://www.sciencedirect.com/science/article/pii/S016786551200219X
Chen, X., Wang, X., Xuan, J.: Tracking multiple moving objects using unscented Kalman filtering techniques. In: International Conference on Engineering and Applied Science (ICEAS 2012), March 2012. https://arxiv.org/abs/1802.01235
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
1 Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Mannepalli, S., Satti, R.V.S.M.R., Shakya, R., Yeturu, K. (2023). Multiple Object Tracking Based on Temporal Local Slice Representation of Sub-regions. In: Gupta, D., Bhurchandi, K., Murala, S., Raman, B., Kumar, S. (eds) Computer Vision and Image Processing. CVIP 2022. Communications in Computer and Information Science, vol 1777. Springer, Cham. https://doi.org/10.1007/978-3-031-31417-9_53
Download citation
DOI: https://doi.org/10.1007/978-3-031-31417-9_53
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-31416-2
Online ISBN: 978-3-031-31417-9
eBook Packages: Computer ScienceComputer Science (R0)