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.
%0 Journal Article
%1 10121650
%A Nguyen, Tuan T.
%A Nguyen, Hoang H.
%A Sartipi, Mina
%A Fisichella, Marco
%D 2024
%J IEEE Transactions on Multimedia
%K #rank1 ITS MTMCT multi-camera myown tracking vehicle
%P 972-983
%R 10.1109/TMM.2023.3274369
%T Multi-Vehicle Multi-Camera Tracking With Graph-Based Tracklet Features
%V 26
%X 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.
@article{10121650,
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.},
added-at = {2024-02-12T17:25:05.000+0100},
author = {Nguyen, Tuan T. and Nguyen, Hoang H. and Sartipi, Mina and Fisichella, Marco},
biburl = {https://www.bibsonomy.org/bibtex/21d59f9b1b65e3a4fbcc8dd6cb3713fdf/erichoang},
doi = {10.1109/TMM.2023.3274369},
interhash = {cacebc30b612312657ad260572a5eb04},
intrahash = {1d59f9b1b65e3a4fbcc8dd6cb3713fdf},
issn = {1941-0077},
journal = {IEEE Transactions on Multimedia},
keywords = {#rank1 ITS MTMCT multi-camera myown tracking vehicle},
pages = {972-983},
timestamp = {2024-02-12T17:25:05.000+0100},
title = {Multi-Vehicle Multi-Camera Tracking With Graph-Based Tracklet Features},
volume = 26,
year = 2024
}