Comprehensive Evaluation of Rule-Based, Machine Learning, and Deep Learning in Human Estimation Using Radio Wave Sensing: Accuracy, Spatial Generalization, and Output Granularity Trade-offs
基于无线电波传感的人类估计的综合评估:规则-based、机器学习与深度学习的准确性、空间泛化与输出粒度的权衡
机构 * School of Electrical and Computer Engineering, Georgia Institute of Technology(电气与计算机工程学院,佐治亚理工学院) ; SoftBank Corp.(软银公司)
AI总结 本研究比较了规则-based、机器学习和深度学习在无线电波传感中的人类估计性能,发现高容量模型在准确性上表现优异但易受领域转移影响,而规则-based方法则更稳健但输出粒度较低。
Comments 10 pages, 5 figures. A comprehensive comparison of rule-based, machine learning, and deep learning approaches for human estimation using FMCW MIMO radar, focusing on accuracy, spatial generalization, and output granularity