arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2110.08595eess.SPcs.CV
收藏

一种基于MIMO雷达的Few-Shot学习方法

A MIMO Radar-based Few-Shot Learning Approach for Human-ID

Fraunhofer IPA(弗劳恩霍夫生产技术与自动化研究所);University of Stuttgart(斯图加特大学)

Pascal Weller, Fady Aziz, Sherif Abdulatif, Urs Schneider, Marco F. Huber

摘要:

用于基于深度学习的人体识别的雷达已成为越来越受关注的研究领域。研究表明,微多普勒($μ$ -D)可以通过捕捉周期性肢体的微运动来反映步行行为。其中一个主要方面是在考虑实时性和训练数据集大小约束的同时,最大化包含的类的数量。本文利用多输入多输出(MIMO)雷达建立了俯仰角速度微动谱($μ$ - $ω$)。研究了将这种新制定的谱图与常用的$μ$ -D相连接的有效性。为了适应无约束的真实行走运动,利用自适应周期分割框架,并在半步态周期($\approx$ 0.5 s)上训练度量学习网络。研究了不同数量的类(5- 20),不同的数据集大小,和不同的观察时间窗口1- 2秒的影响。一个无约束的步行数据集的22个科目收集不同的方面的角度相对于雷达。所提出的Few-Shot学习(FSL)方法仅用每个主题2分钟的训练数据就实现了11.3%的分类错误。

英文摘要:

Radar for deep learning-based human identification has become a research area of increasing interest. It has been shown that micro-Doppler ($μ$-D) can reflect the walking behavior through capturing the periodic limbs' micro-motions. One of the main aspects is maximizing the number of included classes while considering the real-time and training dataset size constraints. In this paper, a multiple-input-multiple-output (MIMO) radar is used to formulate micro-motion spectrograms of the elevation angular velocity ($μ$-$ω$). The effectiveness of concatenating this newly-formulated spectrogram with the commonly used $μ$-D is investigated. To accommodate for non-constrained real walking motion, an adaptive cycle segmentation framework is utilized and a metric learning network is trained on half gait cycles ($\approx$ 0.5 s). Studies on the effects of various numbers of classes (5--20), different dataset sizes, and varying observation time windows 1--2 s are conducted. A non-constrained walking dataset of 22 subjects is collected with different aspect angles with respect to the radar. The proposed few-shot learning (FSL) approach achieves a classification error of 11.3 % with only 2 min of training data per subject.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑