Smart City Gnosys
Smart city article details
| Title | Understanding New Age Of Intelligent Video Surveillance And Deeper Analysis On Deep Learning Techniques For Object Tracking |
|---|---|
| ID_Doc | 59472 |
| Authors | Nagrath P.; Thakur N.; Jain R.; Saini D.; Sharma N.; Hemanth J. |
| Year | 2022 |
| Published | Internet of Things |
| DOI | http://dx.doi.org/10.1007/978-3-030-89554-9_2 |
| Abstract | Surveillance is an imminent part of a smart city model. The persistent possibility of terrorist attacks at public and secured locations raises the need for powerful monitoring systems with subsystems for embedded object tracking. Object tracking is one of machine vision’s basic challenges and has been actively researched for decades. Object tracking is a process to locate a moving object over time across a series of video frames. Object tracking powered with the Internet of Things (IoT) technology provides a broad range of applications such as smart camera surveillance, traffic video surveillance, event prediction and identification, motion detection, human-computer interaction, and perception of human behavior. Real-time visual tracking requires high-response time sensors, tracker speed performance, and large storage requirements. Researchers have ascertained and acknowledged that there is a significant change in the efficacy of drone-based surveillance systems towards object tracking with the inception of the deep learning technologies. Several tracking approaches and models have been proposed by researchers in the area of object tracking and have experienced major improvements with advancement in methods, but object tracking is still considered to be a hard problem to solve. This chapter explains state-of-the-art object tracking algorithms and presents views on current and future trends in object tracking and deep learning surveillance. It also provides an analytical discussion on multi-object tracking experiments based on various datasets available for surveillance and the corresponding results obtained from the research conducted in the near past. FairMOT, GNNMatch, MPNTrack, Lif T, GSDT, and Tracktor++ are among the methods investigated. For the MOT16 and MOT17 datasets, FairMOT generated accuracy of 74.9 and 73.7, respectively, whereas GSDT provided accuracy of 60.7 and 67.1 for the 2DMOT15 and MOT20 datasets. FairMOT is an efficient tracker among the models tested, while MPNTrack is significantly more stable and retains tracklet IDs intact across frames in a series. This concludes FairMOT being an efficient tracker and MPNTrack a stable one. It also discusses a case study on the application of IoT in multi-object tracking and future prospects in surveillance. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG. |
| Author Keywords | Deep learning; Multi-object tracking; Real-time tracking; smart city surveillance; Surveillance |
Similar Articles
| Id | Similarity | Authors | Title | Published |
|---|---|---|---|---|
17814  | 0.889 | Wang X.; Sun Z.; Chehri A.; Jeon G.; Song Y. | Deep Learning And Multi-Modal Fusion For Real-Time Multi-Object Tracking: Algorithms, Challenges, Datasets, And Comparative Study | Information Fusion, 105 (2024) |
| 23882 | 0.863 | Ahmed M.; El-Sheimy N.; Leung H. | Enhancing Object Tracking In Smart City Intelligent Transportation Systems: A Track-By-Detection Approach Utilizing Satellite Video Monitoring | 2024 IEEE International Conference on Smart Mobility, SM 2024 (2024) |
| 34758 | 0.861 | Wang Y.; Zhang L.; Wang X.; Ding K.; Yan J. | Large-Scale Multi-Camera Person Trajectory Tracking Based On Low Sampling Rate Of Camera | Proceedings of the International Conference on Big Data Computing and Communications, BIGCOM, 2024 (2024) |
| 24693 | 0.86 | Kocejko T.; Neumann T.; Mazur-Milecka M.; Kowalczyk N.; Ruminski J.; Jo K.-H.; Kaszynski M.; Ludwisiak T. | Evaluating The Use Of Edge Devices For Detection And Tracking Of Vehicles In Smart City Environment | 2024 International Workshop on Intelligent Systems, IWIS 2024 (2024) |
| 35081 | 0.855 | Jeon Y.; Tran D.Q.; Park M.; Park S. | Leveraging Future Trajectory Prediction For Multi-Camera People Tracking | IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, 2023-June (2023) |
| 21393 | 0.852 | Khochare A.; Sheshadri K.R.; Shriram R.; Simmhan Y. | Dynamic Scaling Of Video Analytics For Wide-Area Tracking In Urban Spaces | Proceedings - 19th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing, CCGrid 2019 (2019) |