Demo: Efficient Convolutional Neural Network for FMCW Radar Based Hand Gesture Recognition

被引：12

作者：

Cai, Xiaodong ^{[1
]}

Ma, Jingyi ^{[2
]}

Liu, Wei ^{[2
]}

Han, Hemin ^{[1
]}

Ma, Lili ^{[1
]}

机构：

[1] Intel Corp, Shanghai, Peoples R China

[2] Intel Labs China, Beijing, Peoples R China

来源：

UBICOMP/ISWC'19 ADJUNCT: PROCEEDINGS OF THE 2019 ACM INTERNATIONAL JOINT CONFERENCE ON PERVASIVE AND UBIQUITOUS COMPUTING AND PROCEEDINGS OF THE 2019 ACM INTERNATIONAL SYMPOSIUM ON WEARABLE COMPUTERS | 2019年

关键词：

FMCW radar; hand gesture recognition; signal processing; Convolutional Neural Network;

D O I：

10.1145/3341162.3343768

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

FMCW radar could detect object's range, speed and Angle-of-Arrival, advantages are robust to bad weather, good range resolution, and good speed resolution. In this paper, we consider the FMCW radar as a novel interacting interface on laptop. We merge sequences of object's range, speed, azimuth information into single input, then feed to a convolution neural network to learn spatial and temporal patterns. Our model achieved 96% accuracy on test set and real-time test.

引用

页码：17 / 20

页数：4

共 6 条

[1]

[Anonymous], IEEE SENSORS LETT

[2] OS-CFAR THEORY FOR MULTIPLE TARGETS AND NONUNIFORM CLUTTER [J].