摘要
提出一种用于实时识别的轻量模型,利用两个子网络分别提取不同类型的特征,在联合训练过程中进行信息共享,并通过知识蒸馏的方式将每个子网络的特征和预测结果融合到一个新的分类器中,用于手势识别.利用循环学习率的训练策略产生了更具泛化性的预测结果.在动态手势数据集DHG-14/28上该方法的识别准确率达到86.11%,验证了该方法的有效性.
For gesture recognition in human-computer interaction,a lightweight model for real-time recognition was proposed.The model consisted of two sub networks,each sub network was responsible for extracting different types of features,realizing a variety of information sharing in the joint training process,and fusing the features and prediction results of each sub network into a new classifier for gesture recognition by knowledge distillation.The cyclic learning rate strategy was adopted to produce more generalized prediction results.Extensive experiments were conducted on the dynamic gesture data set DHG-14/28,demonstrating that the method achieves 86.11%recognition accuracy,which verifies the effectiveness of the method.
作者
许彩芳
XU Caifang(Department of Computer Information,Suzhou Polytechnic College,Suzhou,Anhui 234101,China)
出处
《宜宾学院学报》
2021年第12期31-36,共6页
Journal of Yibin University
基金
安徽省教育厅自然科学重点项目“基于Spark的电商网站用户行为分析预测系统研究”(KJ2019A1058)
安徽省质量工程教研项目“OpenStack云平台部署虚拟仿真实训中心”(2019xfrx06)
安徽省质量工程教研项目“网站开发与网页设计教学团队”(2018jxtd051)。
关键词
深度学习
手势识别
知识共享
知识蒸馏
deep learning
gesture recognition
knowledge sharing
knowledge distillation