Details
| Original language | English |
|---|---|
| Pages (from-to) | 972-983 |
| Number of pages | 12 |
| Journal | IEEE transactions on multimedia |
| Volume | 26 |
| Early online date | 8 May 2023 |
| Publication status | Published - 2024 |
Abstract
Multi-target multi-camera tracking (MTMCT) is an important application in intelligent transportation systems (ITS). The conventional works follow the tracking-by-detection scheme and use the information of the object image separately while matching the object from different cameras. As a result, the association information from the object image is lost. To utilize this information, we propose an efficient MTMCT application that builds features in the form of a graph and customizes graph similarity to match the vehicle objects from different cameras. We present algorithms for both the online scenario, where only the past images are used to match a vehicle object, and the offline scenario, where a given vehicle object is tracked with past and future images. For offline scenarios, our method achieves an IDF1-score of 0.8166 on the Cityflow dataset, which contains the actual scenes of the city from multiple street cameras. For online scenarios, our method achieves an IDF1-score of 0.75 with an FPS of 14.
Keywords
- Cameras, Feature extraction, Graph neural networks, ITS, MTMCT, Multi-Camera Tracking, Object detection, Predictive models, Target tracking, Trajectory, Vehicle Tracking, multi-camera tracking, vehicle tracking
ASJC Scopus subject areas
- Computer Science(all)
- Signal Processing
- Engineering(all)
- Electrical and Electronic Engineering
- Engineering(all)
- Media Technology
- Computer Science(all)
- Computer Science Applications
Sustainable Development Goals
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: IEEE transactions on multimedia, Vol. 26, 2024, p. 972-983.
Research output: Contribution to journal › Article › Research › peer review
}
TY - JOUR
T1 - Multi-Vehicle Multi-Camera Tracking with Graph-Based Tracklet Features
AU - Nguyen, Tuan T.
AU - Nguyen, Hoang H.
AU - Sartipi, Mina
AU - Fisichella, Marco
N1 - ACKNOWLEDGMENT This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes anywarranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof. All authors contributed equally in providing critical feedback and helping shape the research, analysis, planning, and development of the evaluation and manuscript. M.F. conceived the original idea and experimental settings and directed the project. T.T.N. and H.H.N. developed the framework and performed the experiments for the evaluation. M.F. and M.S. verified the analytical methods
PY - 2024
Y1 - 2024
N2 - Multi-target multi-camera tracking (MTMCT) is an important application in intelligent transportation systems (ITS). The conventional works follow the tracking-by-detection scheme and use the information of the object image separately while matching the object from different cameras. As a result, the association information from the object image is lost. To utilize this information, we propose an efficient MTMCT application that builds features in the form of a graph and customizes graph similarity to match the vehicle objects from different cameras. We present algorithms for both the online scenario, where only the past images are used to match a vehicle object, and the offline scenario, where a given vehicle object is tracked with past and future images. For offline scenarios, our method achieves an IDF1-score of 0.8166 on the Cityflow dataset, which contains the actual scenes of the city from multiple street cameras. For online scenarios, our method achieves an IDF1-score of 0.75 with an FPS of 14.
AB - Multi-target multi-camera tracking (MTMCT) is an important application in intelligent transportation systems (ITS). The conventional works follow the tracking-by-detection scheme and use the information of the object image separately while matching the object from different cameras. As a result, the association information from the object image is lost. To utilize this information, we propose an efficient MTMCT application that builds features in the form of a graph and customizes graph similarity to match the vehicle objects from different cameras. We present algorithms for both the online scenario, where only the past images are used to match a vehicle object, and the offline scenario, where a given vehicle object is tracked with past and future images. For offline scenarios, our method achieves an IDF1-score of 0.8166 on the Cityflow dataset, which contains the actual scenes of the city from multiple street cameras. For online scenarios, our method achieves an IDF1-score of 0.75 with an FPS of 14.
KW - Cameras
KW - Feature extraction
KW - Graph neural networks
KW - ITS
KW - MTMCT
KW - Multi-Camera Tracking
KW - Object detection
KW - Predictive models
KW - Target tracking
KW - Trajectory
KW - Vehicle Tracking
KW - multi-camera tracking
KW - vehicle tracking
UR - http://www.scopus.com/inward/record.url?scp=85159826656&partnerID=8YFLogxK
U2 - 10.1109/TMM.2023.3274369
DO - 10.1109/TMM.2023.3274369
M3 - Article
AN - SCOPUS:85159826656
VL - 26
SP - 972
EP - 983
JO - IEEE transactions on multimedia
JF - IEEE transactions on multimedia
SN - 1520-9210
ER -