Abstract
Online tracking is a feature of many state-of-the-art object trackers. When learning online, the data is limited, so the tracker learns a sketch of the object’s features. For a tracker to successfully re-identify the same object in the future frames in many different contexts, including occlusions, the tracker has to keep meta-data over time. In multi-objective inferences, this can exponentially increase the costs and is an ill-posed problem. This paper introduces a model-based framework that combines an ensemble of offline pre-trained models cascaded with domain-specific context for spatial tracking. Our method is efficient in re-identifying objects detected by any camera detector as there is minimal online computation. The second model uses a cosine similarity ranking of the label detected by the first model to find its corresponding set of raw images from the domain training set. A high score means model one has previously seen the object, and a low score amounts to a new detection. By using a two-stage AI-trained ensemble at the edge device, we show that the proposed tracker can perform 10 times faster with its precise detection, and the reidentification at the second stage is accurate, avoiding ID flipping for longer durations on video streams.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Openvino\(^{\rm TM}\) toolkit - open model zoo repository. https://github.com/openvinotoolkit/open_model_zoo. Accessed 30 Aug 2021
Babenko, B., Yang, M.H., Belongie, S.: Visual tracking with online multiple instance learning. In: CVPR (2009)
Bewley, A., Ge, Z., Ott, L., Ramos, F., Upcroft, B.: Simple online and realtime tracking. In: 2016 IEEE International Conference on Image Processing (ICIP), pp. 3464–3468 (2016)
Bolme, D.S., Ross Beveridge, J., Draper, B.A., Lui, Y.M.: Visual object tracking using adaptive correlation filters. In: 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp. 2544–2550 (2010)
Demidovskij, A., et al.: OpenVINO deep learning workbench: towards analytical platform for neural networks inference optimization. In: Journal of Physics: Conference Series, vol. 1828, no. 1, p. 012012, February 2021
Grabner, H., Grabner, M., Bischof, H.: Real-time tracking via on-line boosting. In: Chantler, M.J., Fisher, R.B., Trucco, E. (eds.) BMVC, pp. 47–56. British Machine Vision Association (2006)
He, L., Liao, X., Liu, W., Liu, X., Cheng, P., Mei, T.: FastReID: a pytorch toolbox for general instance re-identification. arXiv preprint arXiv:2006.02631 (2020)
Held, D., Thrun, S., Savarese, S.: Learning to track at 100 FPS with deep regression networks. CoRR arXiv:1604.01802 (2016)
Henriques, J.F., Caseiro, R., Martins, P., Batista, J.: High-speed tracking with kernelized correlation filters. IEEE Trans. Pattern Anal. Mach. Intell. 37(3), 583–596 (2015)
Iyer, V.: Ensemble stream model for data-cleaning in sensor networks. Ph.D. thesis, USA (2013)
Kalal, Z., Mikolajczyk, K., Matas, J.: Tracking-learning-detection. IEEE Trans. Pattern Anal. Mach. Intell. 34(7), 1409–1422 (2012)
Krajewski, R., Bock, J., Kloeker, L., Eckstein, L.: The highD dataset: a drone dataset of naturalistic vehicle trajectories on German highways for validation of highly automated driving systems. In: 2018 21st International Conference on Intelligent Transportation Systems (ITSC), pp. 2118–2125 (2018)
Lam, D., et al.: xView: objects in context in overhead imagery (2018)
Liu, W., Wen, Y., Yu, Z., Yang, M.: Large-margin softmax loss for convolutional neural networks (2017)
Luxonis: DepthAI: embedded machine learning and computer vision API (2020). https://www.luxonis.com/
Luxonis: OAK-D: Stereo camera with edge AI. Stereo Camera with Edge AI capabilites from Luxonis and OpenCV (2020)
Qian, Q., Shang, L., Sun, B., Juhua, H., Li, H., Jin, R.: SoftTriple loss: deep metric learning without triplet sampling (2020)
Zulch, P., Distasio, M., Cushman, T., Wilson, B., Hart, B., Blasch, E.: Escape data collection for multi-modal data fusion research. In: 2019 IEEE Aerospace Conference, pp. 1–10 (2019)
Acknowledgments
We wish to thank JD AI Research’s collaboration during the pandemic on multi-camera with 776 vehicle types captured by 20 cameras, and how spatial OAK-D helped accelerate the prototyping process for understanding spatial edge use-cases. We are grateful for the grant from Google to develop our machine learning course using Google Colab at the undergraduate level. A grant from FIU DOD #3301959-1534-310403, prog20, funded the travel of the first author. First author thanks AFRL mentors for continuing to support object detection using aerial drone images.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Iyer, V., Mehmood, A. (2022). Multi-Object On-Line Tracking as an Ill-Posed Problem: Ensemble Deep Learning at the Edge for Spatial Re-identification. In: Arai, K. (eds) Intelligent Computing. SAI 2022. Lecture Notes in Networks and Systems, vol 507. Springer, Cham. https://doi.org/10.1007/978-3-031-10464-0_13
Download citation
DOI: https://doi.org/10.1007/978-3-031-10464-0_13
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-10463-3
Online ISBN: 978-3-031-10464-0
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)