Staff View: Vision-Based Hand Detection and Tracking Using Fusion of Kernelized Correlation Filter and Single-Shot Detection

Vision-Based Hand Detection and Tracking Using Fusion of Kernelized Correlation Filter and Single-Shot Detection

Hand detection and tracking are key components in many computer vision applications, including hand pose estimation and gesture recognition for human–computer interaction systems, virtual reality, and augmented reality. Despite their importance, reliable hand detection in cluttered scenes remains a...

Full description

Saved in:

Bibliographic Details
Main Authors:	Mohd, Mohd Norzali Haji, Mohd Asaari, Mohd Shahrimie, Ong Lay Ping, Ong Lay Ping, Bakhtiar Affendi Rosdi, Bakhtiar Affendi Rosdi
Format:	Article
Language:	English
Published:	Mdpi 2023
Subjects:	T Technology (General)
Online Access:	http://eprints.uthm.edu.my/9658/1/J16219_8930626b82c06d4375e69da013ec81a8.pdf http://eprints.uthm.edu.my/9658/ https://doi.org/10.3390/app13137433
Tags:	Add Tag No Tags, Be the first to tag this record!

id	my.uthm.eprints.9658
record_format	eprints
spelling	my.uthm.eprints.96582023-08-16T07:11:23Z http://eprints.uthm.edu.my/9658/ Vision-Based Hand Detection and Tracking Using Fusion of Kernelized Correlation Filter and Single-Shot Detection Mohd, Mohd Norzali Haji Mohd Asaari, Mohd Shahrimie Ong Lay Ping, Ong Lay Ping Bakhtiar Affendi Rosdi, Bakhtiar Affendi Rosdi T Technology (General) Hand detection and tracking are key components in many computer vision applications, including hand pose estimation and gesture recognition for human–computer interaction systems, virtual reality, and augmented reality. Despite their importance, reliable hand detection in cluttered scenes remains a challenge. This study explores the use of deep learning techniques for fast and robust hand detection and tracking. A novel algorithm is proposed by combining the Kernelized Correlation Filter (KCF) tracker with the Single-Shot Detection (SSD) method. This integration enables the detection and tracking of hands in challenging environments, such as cluttered backgrounds and occlusions. The SSD algorithm helps reinitialize the KCF tracker when it fails or encounters drift issues due to sudden changes in hand gestures or fast movements. Testing in challenging scenes showed that the proposed tracker achieved a tracking rate of over 90% and a speed of 17 frames per second (FPS). Comparison with the KCF tracker on 17 video sequences revealed an average improvement of 13.31% in tracking detection rate (TRDR) and 27.04% in object detection error (OTE). Additional comparison with MediaPipe hand tracker on 10 hand gesture videos taken from the Intelligent Biometric Group Hand Tracking (IBGHT) dataset showed that the proposed method outperformed the MediaPipe hand tracker in terms of overall TRDR and tracking speed. The results demonstrate the promising potential of the proposed method for long-sequence tracking stability, reducing drift issues, and improving tracking performance during occlusions. Mdpi 2023 Article PeerReviewed text en http://eprints.uthm.edu.my/9658/1/J16219_8930626b82c06d4375e69da013ec81a8.pdf Mohd, Mohd Norzali Haji and Mohd Asaari, Mohd Shahrimie and Ong Lay Ping, Ong Lay Ping and Bakhtiar Affendi Rosdi, Bakhtiar Affendi Rosdi (2023) Vision-Based Hand Detection and Tracking Using Fusion of Kernelized Correlation Filter and Single-Shot Detection. Applied Sciences, 13 (7433). pp. 1-16. https://doi.org/10.3390/app13137433
institution	Universiti Tun Hussein Onn Malaysia
building	UTHM Library
collection	Institutional Repository
continent	Asia
country	Malaysia
content_provider	Universiti Tun Hussein Onn Malaysia
content_source	UTHM Institutional Repository
url_provider	http://eprints.uthm.edu.my/
language	English
topic	T Technology (General)
spellingShingle	T Technology (General) Mohd, Mohd Norzali Haji Mohd Asaari, Mohd Shahrimie Ong Lay Ping, Ong Lay Ping Bakhtiar Affendi Rosdi, Bakhtiar Affendi Rosdi Vision-Based Hand Detection and Tracking Using Fusion of Kernelized Correlation Filter and Single-Shot Detection
description	Hand detection and tracking are key components in many computer vision applications, including hand pose estimation and gesture recognition for human–computer interaction systems, virtual reality, and augmented reality. Despite their importance, reliable hand detection in cluttered scenes remains a challenge. This study explores the use of deep learning techniques for fast and robust hand detection and tracking. A novel algorithm is proposed by combining the Kernelized Correlation Filter (KCF) tracker with the Single-Shot Detection (SSD) method. This integration enables the detection and tracking of hands in challenging environments, such as cluttered backgrounds and occlusions. The SSD algorithm helps reinitialize the KCF tracker when it fails or encounters drift issues due to sudden changes in hand gestures or fast movements. Testing in challenging scenes showed that the proposed tracker achieved a tracking rate of over 90% and a speed of 17 frames per second (FPS). Comparison with the KCF tracker on 17 video sequences revealed an average improvement of 13.31% in tracking detection rate (TRDR) and 27.04% in object detection error (OTE). Additional comparison with MediaPipe hand tracker on 10 hand gesture videos taken from the Intelligent Biometric Group Hand Tracking (IBGHT) dataset showed that the proposed method outperformed the MediaPipe hand tracker in terms of overall TRDR and tracking speed. The results demonstrate the promising potential of the proposed method for long-sequence tracking stability, reducing drift issues, and improving tracking performance during occlusions.
format	Article
author	Mohd, Mohd Norzali Haji Mohd Asaari, Mohd Shahrimie Ong Lay Ping, Ong Lay Ping Bakhtiar Affendi Rosdi, Bakhtiar Affendi Rosdi
author_facet	Mohd, Mohd Norzali Haji Mohd Asaari, Mohd Shahrimie Ong Lay Ping, Ong Lay Ping Bakhtiar Affendi Rosdi, Bakhtiar Affendi Rosdi
author_sort	Mohd, Mohd Norzali Haji
title	Vision-Based Hand Detection and Tracking Using Fusion of Kernelized Correlation Filter and Single-Shot Detection
title_short	Vision-Based Hand Detection and Tracking Using Fusion of Kernelized Correlation Filter and Single-Shot Detection
title_full	Vision-Based Hand Detection and Tracking Using Fusion of Kernelized Correlation Filter and Single-Shot Detection
title_fullStr	Vision-Based Hand Detection and Tracking Using Fusion of Kernelized Correlation Filter and Single-Shot Detection
title_full_unstemmed	Vision-Based Hand Detection and Tracking Using Fusion of Kernelized Correlation Filter and Single-Shot Detection
title_sort	vision-based hand detection and tracking using fusion of kernelized correlation filter and single-shot detection
publisher	Mdpi
publishDate	2023
url	http://eprints.uthm.edu.my/9658/1/J16219_8930626b82c06d4375e69da013ec81a8.pdf http://eprints.uthm.edu.my/9658/ https://doi.org/10.3390/app13137433
_version_	1775624952466112512
score	13.211869

Vision-Based Hand Detection and Tracking Using Fusion of Kernelized Correlation Filter and Single-Shot Detection

Similar Items