Appearance Based Dynamic Hand Gesture Recognition Using 3D Separable Convolutional Neural Network

作     者：Muhammad Rizwan Sana Ul Haq Noor Gul Muhammad Asif Syed Muslim Shah Tariqullah Jan Naveed Ahmad 

作者机构：Department of ElectronicsUniversity of PeshawarPeshawar25120Pakistan Department of Electronics EngineeringKorea Polytechnic UniversitySiheungKorea Department of Electrical EngineeringCapital University of Science and TechnologyIslamabad44000Pakistan Department of Electrical EngineeringUniversity of Engineering and TechnologyPeshawar25120Pakistan Department of Computer SciencePrince Sultan UniversityRiyadh11586Saudi Arabia 

出 版 物：《Computers, Materials & Continua》 (计算机、材料和连续体（英文）)

年 卷 期：2023年第76卷第7期

页      面：1213-1247页

核心收录：

学科分类：0711[理学-系统科学] 07[理学] 08[工学] 081101[工学-控制理论与控制工程] 0811[工学-控制科学与工程] 0701[理学-数学] 0812[工学-计算机科学与技术（可授工学、理学学位）] 071102[理学-系统分析与集成] 081103[工学-系统工程] 

基　　金：Prince Sultan University  PSU 

主　　题：3D separable CNN computational complexity hand gesture recognition human-computer interaction 

摘      要：Appearance-based dynamic Hand Gesture Recognition(HGR)remains a prominent area of research in Human-Computer Interaction(HCI).Numerous environmental and computational constraints limit its real-time *** addition,the performance of a model decreases as the subject’s distance from the camera *** study proposes a 3D separable Convolutional Neural Network(CNN),considering the model’s computa-tional complexity and recognition *** 20BN-Jester dataset was used to train the model for six gesture *** achieving the best offline recognition accuracy of 94.39%,the model was deployed in real-time while considering the subject’s attention,the instant of performing a gesture,and the subject’s distance from the *** being discussed in numerous research articles,the distance factor remains unresolved in real-time deployment,which leads to degraded recognition *** the proposed approach,the distance calculation substantially improves the classification performance by reducing the impact of the subject’s distance from the ***,the capability of feature extraction,degree of relevance,and statistical significance of the proposed model against other state-of-the-art models were validated using t-distributed Stochastic Neighbor Embedding(t-SNE),Mathew’s Correlation Coefficient(MCC),and the McNemar test,*** observed that the proposed model exhibits state-of-the-art outcomes and a comparatively high significance level.