当前位置: X-MOL 学术IEEE Trans. Consum. Electron. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
An Incremental Learning based Gesture Recognition System for Consumer Devices using Edge-Fog Computing
IEEE Transactions on Consumer Electronics ( IF 4.3 ) Pub Date : 2020-02-01 , DOI: 10.1109/tce.2019.2961066
Surbhi Saraswat , Ashish Gupta , Hari Prabhat Gupta , Tanima Dutta

Gesture based systems are attracting more and more researchers to develop a single point control for consumer devices. Most of the existing works use wearable devices or camera based solutions, thus requiring additional resources. This paper presents an incremental learning based gesture recognition system that uses gyroscope sensor of Edge device (mobile phone) to recognize gesture of user and select the function of consumer device. Accelerometer sensor of the mobile phone is then used to control the magnitude of the selected function. User also gives speech input along with the gesture, which is recognized by the Fog device (laptop) and its result is used by the system for incremental learning. The system is implemented by developing an application on the mobile phone and various experiments are performed for validating the accuracy of the system.

中文翻译:

基于增量学习的消费设备手势识别系统使用边缘雾计算

基于手势的系统正在吸引越来越多的研究人员为消费设备开发单点控制。大多数现有作品使用可穿戴设备或基于相机的解决方案,因此需要额外的资源。本文提出了一种基于增量学习的手势识别系统,该系统使用边缘设备(手机)的陀螺仪传感器来识别用户的手势并选择消费设备的功能。然后使用手机的加速度传感器来控制所选功能的大小。用户还提供语音输入以及手势,由 Fog 设备(笔记本电脑)识别,系统使用其结果进行增量学习。
更新日期:2020-02-01
down
wechat
bug