A study on hand gesture recognition algorithm realized with the aid of efficient feature extraction method and convolution neural networks: design and its application to VR environment

Basit Öğe Kaydı
dc.authoridWANG, ZHEN/0000-0003-3927-0115
dc.contributor.authorWang, Zhen
dc.contributor.authorYoo, Sung-Hoon
dc.contributor.authorOh, Sung-Kwun
dc.contributor.authorKim, Eun-Hu
dc.contributor.authorWang, Zheng
dc.contributor.authorFu, Zunwei
dc.contributor.authorJiang, Yuepeng
dc.date.accessioned2024-05-19T14:42:14Z
dc.date.available2024-05-19T14:42:14Z
dc.date.issued2023
dc.departmentİstinye Üniversitesien_US
dc.description.abstractHumans maintain and develop interrelationships through various forms of communication, including verbal and nonverbal communications. Gestures, which constitute one of the most significant forms of nonverbal communication, convey meaning through diverse forms and movements across cultures. In recent decades, research efforts aimed at providing more natural, human-centered means of interacting with computers have garnered increasing interest. Technological advancements in real-time, vision-based hand motion recognition have become progressively suitable for human-computer interaction, aided by computer vision and pattern recognition techniques. Consequently, we propose an effective system for recognizing hand gestures using time-of-flight (ToF) cameras. The hand gesture recognition system outlined in the proposed method incorporates hand shape analysis, as well as robust fingertip and palm center detection. Furthermore, depth sensors, such as ToF cameras, enhance finger detection and hand gesture recognition performance, even in dark or complex backgrounds. Hand shape recognition is performed by comparing newly recognized hand gestures with pre-trained models using a YOLO algorithm-based convolutional neural network. The proposed hand gesture recognition system is implemented in real-world virtual reality applications, and its performance is evaluated based on detection performance and recognition rate outputs. Two distinct gesture recognition datasets, each emphasizing different aspects, were employed. The analysis of results and associated parameters was conducted to evaluate the performance and effectiveness. Experimental results demonstrate that the proposed system achieves competitive classification performance compared to conventional machine learning models evaluated on standard evaluation benchmarks.en_US
dc.description.sponsorshipNational Research Foundation of Korea (NRF) - Korea Government (MSIT) [NRF-2021R1F1A1056102, NRF-2023K2A9A2A0 6060385]; Shandong Excellent Young Scientists Fund Program (Overseas) in China; Taishan Young Scholar Experts Project in China; Basic Science Research Program through the National Research Foundation of Korea (NRF) - Ministry of Education [NRF2022R1I1A1A01071671]en_US
dc.description.sponsorshipThis work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea Government (MSIT) (NRF-2021R1F1A1056102 & amp; NRF-2023K2A9A2A0 6060385), and by Shandong Excellent Young Scientists Fund Program (Overseas) in China and by Taishan Young Scholar Experts Project in China, and also by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF2022R1I1A1A01071671).en_US
dc.identifier.doi10.1007/s00500-023-09077-w
dc.identifier.issn1432-7643
dc.identifier.issn1433-7479
dc.identifier.scopus2-s2.0-85168354284en_US
dc.identifier.scopusqualityQ2en_US
dc.identifier.urihttps://doi.org10.1007/s00500-023-09077-w
dc.identifier.urihttps://hdl.handle.net/20.500.12713/5217
dc.identifier.wosWOS:001051150600001en_US
dc.identifier.wosqualityN/Aen_US
dc.indekslendigikaynakWeb of Scienceen_US
dc.indekslendigikaynakScopusen_US
dc.language.isoenen_US
dc.publisherSpringeren_US
dc.relation.ispartofSoft Computingen_US
dc.relation.publicationcategoryMakale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanıen_US
dc.rightsinfo:eu-repo/semantics/closedAccessen_US
dc.snmz20240519_kaen_US
dc.subjectHand Gesture Recognition (Hrg)en_US
dc.subjectHuman-Computer Interaction (Hci)en_US
dc.subjectTime Of Flight (Tof) Cameraen_US
dc.subjectConvolution Neural Network (Cnn)en_US
dc.subjectVirtual Reality (Vr)en_US
dc.titleA study on hand gesture recognition algorithm realized with the aid of efficient feature extraction method and convolution neural networks: design and its application to VR environmenten_US
dc.typeArticleen_US

Dosyalar

Koleksiyon

WoS İndeksli Yayınlar Koleksiyonu
Scopus İndeksli Yayınlar Koleksiyonu