Shoaib AZMAT, Fazal Wahab KARAM, M. Imran SHEHZAD

Low-cost multiple object tracking for embedded vision applications

This paper presents a low-cost multiple object tracking (MOT) technique by employing a novel appearanceupdate model for object appearance modeling using K-means. The state-of-the-art work has attained a very highaccuracy without considering the real-time aspects necessitated by currently trending embedded vision platforms. Themajor research on multiple object tracking is used to update the appearance model in every frame while discounting itspersistent nature. The proposed appearance update model reduces the computational cost of the state-of-the-art MOT6-fold by exploiting this facet of persistent appearance over the sequence of frames. To ensure accuracy, the proposedmodel is tested on different publicly available standard datasets with challenging situations for both indoor and outdoorscenarios. The experimental results illustrate that our model successfully achieves multiple object tracking while copingwith long-term and complete occlusion. The proposed method achieves the same accuracy in comparison with thestate-of-the-art baseline methods. Moreover, and most importantly, the proposed method is cost-effective in terms ofcomputing and/or memory requirements in comparison to the state-of-the-art techniques. All these traits make ourdesign very suitable for real-time and embedded video surveillance applications with low computing/memory resources.

PDF

___

[1] Ali I, Dailey MN. Multiple human tracking in high-density crowds. Image and Vision Computing 2012; 30 (12): 966-977.
[2] Hongyong T, Youling Y. Finger tracking and gesture recognition with kinect. In: IEEE 2012 International Conference on Computer and Information Technology; Chengdu, China; 2012. pp. 214-218.
[3] Benavidez P, Jamshidi M. Mobile robot navigation and target tracking system. In: IEEE 2011 International Conference on System of Systems Engineering; Albuquerque, NM, USA; 2011. pp. 299-304.
[4] Collins RT, Lipton AJ, Kanade T, Fujiyoshi H, Duggins D et al. A System for Video Surveillance and Monitoring. Pittsburgh, PA, USA: Carnegie Mellon University Press, 2000.
[5] Haritaoglu I, Harwood D, Davis LS. W/sup 4/: real-time surveillance of people and their activities. IEEE Transactions on Pattern Analysis & Machine Intelligence 2000; 22 (8): 809-830.
[6] Stauffer C, Grimson WE. Learning patterns of activity using real-time tracking. IEEE Transactions on Pattern Analysis & Machine Intelligence 2000; 22(8): 747-757.
[7] Führ G, Jung CR. Combining patch matching and detection for robust pedestrian tracking in monocular calibrated cameras. Pattern Recognition Letters 2014; 39: 11-20.
[8] Bae SH, Yoon KJ. Robust online multi-object tracking based on tracklet confidence and online discriminative appearance learning. In: IEEE 2014 Conference on Computer Vision and Pattern Recognition; Columbus, OH, USA; 2014. pp. 1218-1225.
[9] Milan A, Leal-Taixé L, Reid I, Roth S, Schindler K. MOT16: A Benchmark for Multi-object Tracking. Ithaca, NY, USA: Cornell University Repository, 2016.
[10] Dalal N, Triggs B. Histograms of oriented gradients for human detection. In: IEEE 2005 Conference on Computer Vision and Pattern Recognition; San Diego, CA, USA; 2005. pp. 886-893.
[11] Breitenstein MD, Reichlin F, Leibe B, Koller-Meier E, Van Gool L. Robust tracking-by-detection using a detector confidence particle filter. In: IEEE 2009 International Conference on Computer Vision; Kyoto, Japan; 2009. pp. 1515-1522.
[12] Gaikwad V, Lokhande S. Vision based pedestrian detection for advanced driver assistance. Procedia Computer Science 2015; 46: 321-8.
[13] Lim J, Kim W. Detecting and tracking of multiple pedestrians using motion, color information and the AdaBoost algorithm. Multimedia Tools and Applications 2013; 65 (1): 161-79.
[14] Milan A, Rezatofighi SH, Dick AR, Reid ID, Schindler K. Online multi-target tracking using recurrent neural networks. In: AAAI 2017 Conference on Artificial Intelligence; San Francisco, CA, USA; 2017. pp. 4225–4232.
[15] Sadeghian A, Alahi A, Savarese S. Tracking the untrackable: learning to track multiple cues with long-term dependencies. In: IEEE 2017 International Conference on Computer Vision; Venice, Italy; 2017. pp. 300–311.
[16] Gan W, Wang S, Lei X, Lee MS, Kuo CC. Online CNN-based multiple object tracking with enhanced model updates and identity association. Signal Processing Image Communication 2018; 66: 95-102.
[17] Chu Q, Ouyang W, Li H, Wang X, Liu B et al. Online multi-object tracking using CNN based single object tracker with spatial-temporal attention mechanism. In: IEEE 2017 International Conference on Computer Vision; Venice, Italy; 2017. pp. 4846-4855.
[18] Nguyen HT, Smeulders AW. Fast occluded object tracking by a robust appearance filter. IEEE Transactions on Pattern Analysis and Machine Intelligence 2004; 26 (8): 1099-1104.
[19] Luo W, Xing J, Milan A, Zhang X, Liu W et al. Multiple Object Tracking: A Literature Review. Ithaca, NY, USA: Cornell University Repository, 2014.
[20] Kratz L, Nishino K. Tracking pedestrians using local spatio-temporal motion patterns in extremely crowded scenes. IEEE Transactions on Pattern Analysis and Machine Intelligence 2012; 34 (5): 987-1002.
[21] Yang T, Pan Q, Li J, Li S. Real-time multiple objects tracking with occlusion handling in dynamic scenes. In: IEEE 2005 International Conference on Computer Vision and Pattern Recognition; San Diego, CA, USA; 2005. pp.
[22] Henriques JF, Caseiro R, Batista J. Globally optimal solution to multi-object tracking with merged measurements. In: IEEE 2011 International Conference on Computer Vision and Pattern Recognition; Barcelona, Spain; 2011. pp. 2470-2477.
[23] Tao H, Sawhney HS, Kumar R. Object tracking with Bayesian estimation of dynamic layer representations. IEEE Transactions on Pattern Analysis and Machine Intelligence 2002; 24 (1): 75-89.
[24] Khan S, Shah M. Tracking people in presence of occlusion. In: Springer 2000 Asian Conference on Computer Vision; Taipei, Taiwan; 2000. pp. 1132-1137.
[25] Papadourakis V, Argyros A. Multiple objects tracking in the presence of long-term occlusions. Computer Vision and Image Understanding 2010; 114 (7): 835-846.
[26] Shehzad MI, Shah YA, Mehmood Z, Malik AW, Azmat S. K-means based multiple objects tracking with long-term occlusion handling. IET Computer Vision 2017; 11 (1): 68-77.
[27] Leal-Taixé L, Milan A, Reid I, Roth S, Schindler K. Motchallenge 2015: Towards a Benchmark for Multi-target Tracking. Ithaca, NY, USA: Cornell University Repository, 2015.
[28] Pandey M, Ubhi JS, Raju KS. Computational acceleration of real-time kernel-based tracking system. Journal of Circuits, Systems and Computers 2016; 25 (4): 1-30.
[29] Sanchez-Matilla R, Poiesi F, Cavallaro A. Online multi-target tracking with strong and weak detections. In: Springer 2016 European Conference on Computer Vision; Amsterdam, the Netherlands; 2016. pp. 84-99.
[30] Song YM, Jeon M. Online multiple object tracking with the hierarchically adopted gm-phd filter using motion and appearance. In: IEEE 2016 International Conference on Consumer Electronics-Asia; Seoul, South Korea; 2016. pp. 1-4.
[31] Ju J, Kim D, Ku B, Han DK, Ko H. Online multi-person tracking with two-stage data association and online appearance model learning. IET Computer Vision 2016; 11 (1): 87-95.
[32] Apewokin S, Valentine B, Wills L, Wills S, Gentile A. Multimodal mean adaptive background for embedded realtime video surveillance. In: IEEE 2007 Conference on Computer Vision and Pattern Recognition; Minneapolis, MN, USA; 2007. pp. 1-6.
[33] Tian M, Yang Q, Maier A, Schasiepen I, Maass N et al. Automatic histogram-based initialization of k-means clustering in CT. In: Springer 2013 Workshop Bildverarbeitung fur die Medizin; Heidelberg, Germany; 2013. pp. 277-282.
[34] Li X, Hu W, Shen C, Zhang Z, Dick A et al. A survey of appearance models in visual object tracking. ACM Transactions on Intelligent Systems and Technology 2013; 4 (4): 1-48.
[35] Barry B, Brick C, Connor F, Donohoe D, Moloney D et al. Always-on vision processing unit for mobile applications. IEEE Micro 2015; 35 (2): 56-66.