Comparison of Four SVM Classifiers Used with Depth Sensors to Recognize Arabic Sign Language Words

Almasre, Miada A.; Al-Nuaim, Hana

doi:10.3390/computers6020020

Cited by 11 publications

(8 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For classification, a modified K-Nearest Neighbor (MKNN) is used and achieved an accuracy of 98.9%. Depth sensors were used to capture the upper part of the body to recognize hand gestures showing Arabic sign language [41]. The dataset consists of five words with 143 signs examined by ten people.…”

Section: Related Workmentioning

confidence: 99%

Arabic and American Sign Languages Alphabet Recognition by Convolutional Neural Network

Alshomrani¹,

Al-joudi²,

Arif³

2021

Adv. Sci. Technol. Res. J.

View full text Add to dashboard Cite

Section: Related Workmentioning

confidence: 99%

Arabic and American Sign Languages Alphabet Recognition by Convolutional Neural Network

Alshomrani¹,

Al-joudi²,

Arif³

2021

Adv. Sci. Technol. Res. J.

View full text Add to dashboard Cite

“…Kinect can work only with Windows operating systems. Windows Kinect version 2.0 allows developers to create applications that need voice recognition, depth camera or face-tracking to capture the movement of the human body" as explained in [12]. The use of this sensor will allow the system to recognize any gesture/movement made by a human from a reasonable distance.…”

Section: A Hardware Set-up and Installationmentioning

confidence: 99%

HoloLearn: An Interactive Educational System

Alghamdi¹,

Aloufi²,

Alsalea³

et al. 2021

IJACSA

View full text Add to dashboard Cite

The HoloLearn project is a sophisticated interactive educational system that attempts to simplify the educational process in the field of medicine, mainly, through the use of Hologram technology. The Hologram technology has been used in conjunction with the feature of user interaction, to take the whole educational process to a completely new level providing students with a different learning experience. The system is more dedicated to medical students as they must study diverse, complicated structures of the human body anatomy and its internal organs. HoloLearn is aimed at replacing the traditional educational techniques with a new one that involves user interaction with real-sized 3D objects. Based on the interview conducted with medical students from different universities and educational levels in Saudi Arabia, and the questionnaire results, it has been found that traditional learning techniques are insufficient/inefficient as they lack quality and most of the criteria that could qualify them to be highly effective as reliable learning materials. Therefore, there is an increasing need for new learning strategies/methods with enough capabilities to give the students the chance to perceive every concept they study, rather than depending on their imagination to picture what a human body looks like from inside, they need visual learning methods. From another perspective, teachers also face difficulties when explaining medical concepts, especially those related to human body structure and behavior. The currently available materials and sources are mostly theoretical. They promote indoctrination and a result-driven approach instead of engaging the students in a process of sharing knowledge and ideas from both parties i.e., teacher and students. In fact, students have to listen and read instead of practicing and exploring, consequently students are prone to loss of concentration and mental distraction during lessons repeatedly, while from another angle they suffer from long study hours and difficulties in retrieving information. The results of this project indicate that when combining the Hologram technology with the user interaction feature, the educational process can be highly improved, and can be much more creative and entertaining.

show abstract

“…Advances in 3D depth cameras that were utilized by Kinect have produced many opportunities for multimedia computing [15]. Kinect relies on depth technology, which allows users to deal with any system via a web camera [2,16].…”

Section: Sensorsmentioning

confidence: 99%

“…It is a vital step to determine which classification algorithm is suitable to represent the data, and how well this chosen algorithm will work in the future (Sayad, 2010). Statistical models that are based on the values of Accuracy, Area Under Curve (AUC), and logarithmic loss (logLoss) can be used to evaluate the model [16].…”

Section: Classificationmentioning

confidence: 99%

A comparison of Arabic sign language dynamic gesture recognition models

Almasre

Al-Nuaim

2020

Heliyon

Self Cite

View full text Add to dashboard Cite

Arabic Sign Language (ArSL) is similar to other sign languages in terms of the way it is gestured and interpreted and used as a medium of communication among the hearing-impaired and the communities in which they live in. Research investigating sensor utilization and natural user interfaces to facilitate ArSL recognition and interpretation, is lacking.Previous research has demonstrated that there is not a single classifier modeling approach that can be suitable for all hand gesture recognition tasks, therefore, this research investigated which combination of algorithms, set with different parameters used with a sensor device, produce higher ArSL recognition accuracy results in a gesture recognition system.This research proposed a dynamic prototype model (DPM) using Kinect as a sensor to recognize certain ArSL gestured dynamic words. The DPM used eleven predictive models of three algorithms (SVM, RF, KNN) based on different parameter settings. Research findings indicated that highest recognition accuracy rates for the dynamic words gestured were achieved by the SVM models, with linear kernel and cost parameter ¼ 0.035.

show abstract

Comparison of Four SVM Classifiers Used with Depth Sensors to Recognize Arabic Sign Language Words

Cited by 11 publications

References 18 publications

Arabic and American Sign Languages Alphabet Recognition by Convolutional Neural Network

Arabic and American Sign Languages Alphabet Recognition by Convolutional Neural Network

HoloLearn: An Interactive Educational System

A comparison of Arabic sign language dynamic gesture recognition models

Contact Info

Product

Resources

About