.png){kind=icon}
返回
A fine-grained gesture tracking system based on millimeter-wave
.png){kind=icon}
PRE
AI
DOI：10.1007/s42486-022-00119-0.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
As an essential part of modern human-computer interaction, gesture recognition is widely used in industry, society, medical care and entertainment. Existing gesture recognition solutions either rely on computer vision, which suffers from the light condition, or use inertial sensors, which are limited by the battery life. In this paper, we propose a millimeter-wave based solution to recognize the human gesture for natural interaction, which can efficiently solve the above limitations. We leverage a frequency-modulated continuous wave radar to implement a cm-level fine-grained gesture tracking system. In order to obtain the target position, we build a theoretical model between signals and position by applying a multi-dimensional fast Fourier transform to the captured data. To handle the interference of the multipath effect, we propose a distance-based voting method to delete unreasonable points and optimize the trajectory. Furthermore, we design an optimal estimation algorithm based on measurements with different precision to obtain more acceptable estimation. Finally, in view of the sparse points, we propose an interpolation method of fitting traces in different dimensions to augment the point cloud. We have implemented a prototype system, and the experimental results show that our system can accurately track the gestures without knowing the gesture in advance. The average trajectory position error is 0.94 cm, and the average normalized location error is 0.027 for gestures with the size of about 20 cm x40 cm.
Keyword:
Millimeter-wave
Gesture recognition
Echo data processing
Tracking
期刊
C
IF:
2
论文数:
35
被引数:
296
机构
引用论文
.png){kind=icon}
分享收藏
查看更多内容