Vulnerable Road Users Detection and Tracking using YOLOv4 and Deep SORT

Ahmed H. Abdel-Gawad; Alaa Khamis; Lobna A. Said; Ahmed G. Radwan

doi:10.1109/JAC-ECC54461.2021.9691441

Vulnerable Road Users Detection and Tracking using YOLOv4 and Deep SORT

Ahmed H. Abdel-Gawad, Alaa Khamis, Lobna A. Said, Ahmed G. Radwan

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

4 Scopus citations

Abstract

Over the years, The detection and tracking of Vulnerable Road Users (VRUs) have become one of the most critical features of self-driving car components. Because of its processing efficiency and better detection algorithms, tracking-by-detection appears to be the best paradigm. In this paper, a detection-based tracking approach is presented for Multiple VRU Tracking of video from an inside-vehicle camera in real-time. YOLOv4 scans every frame to detect VRUs first, then Simple Online and Realtime Tracking with a Deep Association Metric (Deep SORT) algorithm, which is customized for multiple VRU tracking, is applied. The results of our experiments on both the Joint Attention in Autonomous Driving (JAAD) and Multiple Object Tracking (MOT) datasets exhibit competitive performance.

Original language	English
Title of host publication	Proceedings of the 2021 International Japan-Africa Conference on Electronics, Communications, and Computations, JAC-ECC 2021
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	140-145
Number of pages	6
ISBN (Electronic)	9781665482929
DOIs	https://doi.org/10.1109/JAC-ECC54461.2021.9691441
State	Published - 2021
Externally published	Yes
Event	9th International Japan-Africa Conference on Electronics, Communications, and Computations, JAC-ECC 2021 - Virtual, Online, Egypt Duration: 13 Dec 2021 → 14 Dec 2021

Publication series

Name	Proceedings of the 2021 International Japan-Africa Conference on Electronics, Communications, and Computations, JAC-ECC 2021

Conference

Conference	9th International Japan-Africa Conference on Electronics, Communications, and Computations, JAC-ECC 2021
Country/Territory	Egypt
City	Virtual, Online
Period	13/12/21 → 14/12/21

Bibliographical note

Publisher Copyright:
© 2021 IEEE.

Keywords

Computer Vision
Deep SORT
Object Detection
Object Tracking
Self-Driving Vehicles
Vulnerable Road User (VRU)
YOLOv4

ASJC Scopus subject areas

Computer Science Applications
Computer Networks and Communications
Hardware and Architecture
Electrical and Electronic Engineering
Instrumentation

Access to Document

10.1109/JAC-ECC54461.2021.9691441

Cite this

Abdel-Gawad, A. H., Khamis, A., Said, L. A., & Radwan, A. G. (2021). Vulnerable Road Users Detection and Tracking using YOLOv4 and Deep SORT. In Proceedings of the 2021 International Japan-Africa Conference on Electronics, Communications, and Computations, JAC-ECC 2021 (pp. 140-145). (Proceedings of the 2021 International Japan-Africa Conference on Electronics, Communications, and Computations, JAC-ECC 2021). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/JAC-ECC54461.2021.9691441

Abdel-Gawad, Ahmed H. ; Khamis, Alaa ; Said, Lobna A. et al. / Vulnerable Road Users Detection and Tracking using YOLOv4 and Deep SORT. Proceedings of the 2021 International Japan-Africa Conference on Electronics, Communications, and Computations, JAC-ECC 2021. Institute of Electrical and Electronics Engineers Inc., 2021. pp. 140-145 (Proceedings of the 2021 International Japan-Africa Conference on Electronics, Communications, and Computations, JAC-ECC 2021).

@inproceedings{63856b17bd1a40959d456e9e2a5976b6,

title = "Vulnerable Road Users Detection and Tracking using YOLOv4 and Deep SORT",

abstract = "Over the years, The detection and tracking of Vulnerable Road Users (VRUs) have become one of the most critical features of self-driving car components. Because of its processing efficiency and better detection algorithms, tracking-by-detection appears to be the best paradigm. In this paper, a detection-based tracking approach is presented for Multiple VRU Tracking of video from an inside-vehicle camera in real-time. YOLOv4 scans every frame to detect VRUs first, then Simple Online and Realtime Tracking with a Deep Association Metric (Deep SORT) algorithm, which is customized for multiple VRU tracking, is applied. The results of our experiments on both the Joint Attention in Autonomous Driving (JAAD) and Multiple Object Tracking (MOT) datasets exhibit competitive performance.",

keywords = "Computer Vision, Deep SORT, Object Detection, Object Tracking, Self-Driving Vehicles, Vulnerable Road User (VRU), YOLOv4",

author = "Abdel-Gawad, \{Ahmed H.\} and Alaa Khamis and Said, \{Lobna A.\} and Radwan, \{Ahmed G.\}",

note = "Publisher Copyright: {\textcopyright} 2021 IEEE.; 9th International Japan-Africa Conference on Electronics, Communications, and Computations, JAC-ECC 2021 ; Conference date: 13-12-2021 Through 14-12-2021",

year = "2021",

doi = "10.1109/JAC-ECC54461.2021.9691441",

language = "English",

series = "Proceedings of the 2021 International Japan-Africa Conference on Electronics, Communications, and Computations, JAC-ECC 2021",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "140--145",

booktitle = "Proceedings of the 2021 International Japan-Africa Conference on Electronics, Communications, and Computations, JAC-ECC 2021",

address = "United States",

}

Abdel-Gawad, AH, Khamis, A, Said, LA & Radwan, AG 2021, Vulnerable Road Users Detection and Tracking using YOLOv4 and Deep SORT. in Proceedings of the 2021 International Japan-Africa Conference on Electronics, Communications, and Computations, JAC-ECC 2021. Proceedings of the 2021 International Japan-Africa Conference on Electronics, Communications, and Computations, JAC-ECC 2021, Institute of Electrical and Electronics Engineers Inc., pp. 140-145, 9th International Japan-Africa Conference on Electronics, Communications, and Computations, JAC-ECC 2021, Virtual, Online, Egypt, 13/12/21. https://doi.org/10.1109/JAC-ECC54461.2021.9691441

Vulnerable Road Users Detection and Tracking using YOLOv4 and Deep SORT. / Abdel-Gawad, Ahmed H.; Khamis, Alaa; Said, Lobna A. et al.
Proceedings of the 2021 International Japan-Africa Conference on Electronics, Communications, and Computations, JAC-ECC 2021. Institute of Electrical and Electronics Engineers Inc., 2021. p. 140-145 (Proceedings of the 2021 International Japan-Africa Conference on Electronics, Communications, and Computations, JAC-ECC 2021).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Vulnerable Road Users Detection and Tracking using YOLOv4 and Deep SORT

AU - Abdel-Gawad, Ahmed H.

AU - Khamis, Alaa

AU - Said, Lobna A.

AU - Radwan, Ahmed G.

PY - 2021

Y1 - 2021

N2 - Over the years, The detection and tracking of Vulnerable Road Users (VRUs) have become one of the most critical features of self-driving car components. Because of its processing efficiency and better detection algorithms, tracking-by-detection appears to be the best paradigm. In this paper, a detection-based tracking approach is presented for Multiple VRU Tracking of video from an inside-vehicle camera in real-time. YOLOv4 scans every frame to detect VRUs first, then Simple Online and Realtime Tracking with a Deep Association Metric (Deep SORT) algorithm, which is customized for multiple VRU tracking, is applied. The results of our experiments on both the Joint Attention in Autonomous Driving (JAAD) and Multiple Object Tracking (MOT) datasets exhibit competitive performance.

AB - Over the years, The detection and tracking of Vulnerable Road Users (VRUs) have become one of the most critical features of self-driving car components. Because of its processing efficiency and better detection algorithms, tracking-by-detection appears to be the best paradigm. In this paper, a detection-based tracking approach is presented for Multiple VRU Tracking of video from an inside-vehicle camera in real-time. YOLOv4 scans every frame to detect VRUs first, then Simple Online and Realtime Tracking with a Deep Association Metric (Deep SORT) algorithm, which is customized for multiple VRU tracking, is applied. The results of our experiments on both the Joint Attention in Autonomous Driving (JAAD) and Multiple Object Tracking (MOT) datasets exhibit competitive performance.

KW - Computer Vision

KW - Deep SORT

KW - Object Detection

KW - Object Tracking

KW - Self-Driving Vehicles

KW - Vulnerable Road User (VRU)

KW - YOLOv4

UR - https://www.scopus.com/pages/publications/85126801558

U2 - 10.1109/JAC-ECC54461.2021.9691441

DO - 10.1109/JAC-ECC54461.2021.9691441

M3 - Conference contribution

AN - SCOPUS:85126801558

T3 - Proceedings of the 2021 International Japan-Africa Conference on Electronics, Communications, and Computations, JAC-ECC 2021

SP - 140

EP - 145

BT - Proceedings of the 2021 International Japan-Africa Conference on Electronics, Communications, and Computations, JAC-ECC 2021

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 9th International Japan-Africa Conference on Electronics, Communications, and Computations, JAC-ECC 2021

Y2 - 13 December 2021 through 14 December 2021

ER -

Abdel-Gawad AH, Khamis A, Said LA, Radwan AG. Vulnerable Road Users Detection and Tracking using YOLOv4 and Deep SORT. In Proceedings of the 2021 International Japan-Africa Conference on Electronics, Communications, and Computations, JAC-ECC 2021. Institute of Electrical and Electronics Engineers Inc. 2021. p. 140-145. (Proceedings of the 2021 International Japan-Africa Conference on Electronics, Communications, and Computations, JAC-ECC 2021). doi: 10.1109/JAC-ECC54461.2021.9691441

Vulnerable Road Users Detection and Tracking using YOLOv4 and Deep SORT

Abstract

Publication series

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this