本文已被:浏览 98次 下载 88次
投稿时间:2018-05-02 修订日期:2018-05-24
投稿时间:2018-05-02 修订日期:2018-05-24
中文摘要: 本文是对SKIG RGB-D多模态的孤立手势视频进行手势识别研究.首先将RGB和Depth两种单模态视频提取成图片的形式保存,然后采样成长度为32帧的手势序列分别输入到本文提出的稠密连接的3DCNN组件学习短期的时空域特征,然后将提取的时空域特征输入到卷积GRU网络进行长期的时空域特征学习,最终对单模态训练好的网络进行多模态融合,提升网络识别准确率.本文在SKIG数据集上取得了99.07%的识别准确率,达到了极高的准确率,证明了本文提出的网络模型的有效性.
中文关键词: 手势识别 稠密连接的3DCNN 卷积GRU 时空域特征
Abstract:In this study, the gesture recognition based on SKIG RGB-D multimodal isolated gesture video is studied. The RGB and depth videos are extracted into the form of images. Then the sampled 32 frames from images are input to the densely connected 3DCNN component to learn short-term spatiotemporal features, after that the features input to the convolutional GRU to learn long-term spatiotemporal features. Finally, the trained networks for single modal are used to multimodal fusion to improve the recognition accuracy. 99.07% recognition accuracy is obtained on the SKIG dataset, which achieves high accuracy and proves the validity of the network model proposed in this study.
文章编号: 中图分类号: 文献标志码:
基金项目:福建省自然科学基金(2017J01744)
引用文本:
马正文,蔡坚勇,刘磊,欧阳乐峰,李楠.基于RGB-D视频的多模态手势识别.计算机系统应用,2018,27(12):234-239
MA Zheng-Wen,CAI Jian-Yong,LIU Lei,OUYANG Le-Feng,LI Nan.Multimodal Gesture Recognition Based on RGB-D Video.COMPUTER SYSTEMS APPLICATIONS,2018,27(12):234-239
马正文,蔡坚勇,刘磊,欧阳乐峰,李楠.基于RGB-D视频的多模态手势识别.计算机系统应用,2018,27(12):234-239
MA Zheng-Wen,CAI Jian-Yong,LIU Lei,OUYANG Le-Feng,LI Nan.Multimodal Gesture Recognition Based on RGB-D Video.COMPUTER SYSTEMS APPLICATIONS,2018,27(12):234-239