Systematized event-aware learning for multi-object tracking

Hyemin Lee, Daijin Kim
Proceedings of the Thirty-Eighth Conference on Uncertainty in Artificial Intelligence, PMLR 180:1074-1084, 2022.

Abstract

We propose an end-to-end online multi-object tracking (MOT) framework with a systematized event-aware loss, which is designed to control possible occurrences in an online MOT situation and compel the tracker to take appropriate actions when such events occur. Training samples from real candidates using a simulation tracker are generated, and a systematized event-aware association matrix is constructed for every frame to enable the tracker to learn the ideal action in a running environment. Several experiments, including ablation studies on various public MOT benchmark datasets, are conducted. The experimental results verify that each event affecting the tracking measure can be controlled, and the proposed method presents optimal results compared with recent state-of-the-art MOT methods.

Cite this Paper

BibTeX
@InProceedings{pmlr-v180-lee22a, title = {Systematized event-aware learning for multi-object tracking}, author = {Lee, Hyemin and Kim, Daijin}, booktitle = {Proceedings of the Thirty-Eighth Conference on Uncertainty in Artificial Intelligence}, pages = {1074--1084}, year = {2022}, editor = {Cussens, James and Zhang, Kun}, volume = {180}, series = {Proceedings of Machine Learning Research}, month = {01--05 Aug}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v180/lee22a/lee22a.pdf}, url = {https://proceedings.mlr.press/v180/lee22a.html}, abstract = {We propose an end-to-end online multi-object tracking (MOT) framework with a systematized event-aware loss, which is designed to control possible occurrences in an online MOT situation and compel the tracker to take appropriate actions when such events occur. Training samples from real candidates using a simulation tracker are generated, and a systematized event-aware association matrix is constructed for every frame to enable the tracker to learn the ideal action in a running environment. Several experiments, including ablation studies on various public MOT benchmark datasets, are conducted. The experimental results verify that each event affecting the tracking measure can be controlled, and the proposed method presents optimal results compared with recent state-of-the-art MOT methods.} }
Endnote
%0 Conference Paper %T Systematized event-aware learning for multi-object tracking %A Hyemin Lee %A Daijin Kim %B Proceedings of the Thirty-Eighth Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2022 %E James Cussens %E Kun Zhang %F pmlr-v180-lee22a %I PMLR %P 1074--1084 %U https://proceedings.mlr.press/v180/lee22a.html %V 180 %X We propose an end-to-end online multi-object tracking (MOT) framework with a systematized event-aware loss, which is designed to control possible occurrences in an online MOT situation and compel the tracker to take appropriate actions when such events occur. Training samples from real candidates using a simulation tracker are generated, and a systematized event-aware association matrix is constructed for every frame to enable the tracker to learn the ideal action in a running environment. Several experiments, including ablation studies on various public MOT benchmark datasets, are conducted. The experimental results verify that each event affecting the tracking measure can be controlled, and the proposed method presents optimal results compared with recent state-of-the-art MOT methods.
APA
Lee, H. & Kim, D.. (2022). Systematized event-aware learning for multi-object tracking. Proceedings of the Thirty-Eighth Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research 180:1074-1084 Available from https://proceedings.mlr.press/v180/lee22a.html.

Related Material