Cargando…

Efficient Single-Shot Multi-Object Tracking for Vehicles in Traffic Scenarios

Multi-object tracking is a significant field in computer vision since it provides essential information for video surveillance and analysis. Several different deep learning-based approaches have been developed to improve the performance of multi-object tracking by applying the most accurate and effi...

Descripción completa

Detalles Bibliográficos
Autores principales: Lee, Youngkeun, Lee, Sang-ha, Yoo, Jisang, Kwon, Soonchul
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2021
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8512362/
https://www.ncbi.nlm.nih.gov/pubmed/34640675
http://dx.doi.org/10.3390/s21196358
_version_ 1784582972623552512
author Lee, Youngkeun
Lee, Sang-ha
Yoo, Jisang
Kwon, Soonchul
author_facet Lee, Youngkeun
Lee, Sang-ha
Yoo, Jisang
Kwon, Soonchul
author_sort Lee, Youngkeun
collection PubMed
description Multi-object tracking is a significant field in computer vision since it provides essential information for video surveillance and analysis. Several different deep learning-based approaches have been developed to improve the performance of multi-object tracking by applying the most accurate and efficient combinations of object detection models and appearance embedding extraction models. However, two-stage methods show a low inference speed since the embedding extraction can only be performed at the end of the object detection. To alleviate this problem, single-shot methods, which simultaneously perform object detection and embedding extraction, have been developed and have drastically improved the inference speed. However, there is a trade-off between accuracy and efficiency. Therefore, this study proposes an enhanced single-shot multi-object tracking system that displays improved accuracy while maintaining a high inference speed. With a strong feature extraction and fusion, the object detection of our model achieves an AP score of 69.93% on the UA-DETRAC dataset and outperforms previous state-of-the-art methods, such as FairMOT and JDE. Based on the improved object detection performance, our multi-object tracking system achieves a MOTA score of 68.5% and a PR-MOTA score of 24.5% on the same dataset, also surpassing the previous state-of-the-art trackers.
format Online
Article
Text
id pubmed-8512362
institution National Center for Biotechnology Information
language English
publishDate 2021
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-85123622021-10-14 Efficient Single-Shot Multi-Object Tracking for Vehicles in Traffic Scenarios Lee, Youngkeun Lee, Sang-ha Yoo, Jisang Kwon, Soonchul Sensors (Basel) Article Multi-object tracking is a significant field in computer vision since it provides essential information for video surveillance and analysis. Several different deep learning-based approaches have been developed to improve the performance of multi-object tracking by applying the most accurate and efficient combinations of object detection models and appearance embedding extraction models. However, two-stage methods show a low inference speed since the embedding extraction can only be performed at the end of the object detection. To alleviate this problem, single-shot methods, which simultaneously perform object detection and embedding extraction, have been developed and have drastically improved the inference speed. However, there is a trade-off between accuracy and efficiency. Therefore, this study proposes an enhanced single-shot multi-object tracking system that displays improved accuracy while maintaining a high inference speed. With a strong feature extraction and fusion, the object detection of our model achieves an AP score of 69.93% on the UA-DETRAC dataset and outperforms previous state-of-the-art methods, such as FairMOT and JDE. Based on the improved object detection performance, our multi-object tracking system achieves a MOTA score of 68.5% and a PR-MOTA score of 24.5% on the same dataset, also surpassing the previous state-of-the-art trackers. MDPI 2021-09-23 /pmc/articles/PMC8512362/ /pubmed/34640675 http://dx.doi.org/10.3390/s21196358 Text en © 2021 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Lee, Youngkeun
Lee, Sang-ha
Yoo, Jisang
Kwon, Soonchul
Efficient Single-Shot Multi-Object Tracking for Vehicles in Traffic Scenarios
title Efficient Single-Shot Multi-Object Tracking for Vehicles in Traffic Scenarios
title_full Efficient Single-Shot Multi-Object Tracking for Vehicles in Traffic Scenarios
title_fullStr Efficient Single-Shot Multi-Object Tracking for Vehicles in Traffic Scenarios
title_full_unstemmed Efficient Single-Shot Multi-Object Tracking for Vehicles in Traffic Scenarios
title_short Efficient Single-Shot Multi-Object Tracking for Vehicles in Traffic Scenarios
title_sort efficient single-shot multi-object tracking for vehicles in traffic scenarios
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8512362/
https://www.ncbi.nlm.nih.gov/pubmed/34640675
http://dx.doi.org/10.3390/s21196358
work_keys_str_mv AT leeyoungkeun efficientsingleshotmultiobjecttrackingforvehiclesintrafficscenarios
AT leesangha efficientsingleshotmultiobjecttrackingforvehiclesintrafficscenarios
AT yoojisang efficientsingleshotmultiobjecttrackingforvehiclesintrafficscenarios
AT kwonsoonchul efficientsingleshotmultiobjecttrackingforvehiclesintrafficscenarios