Fast and scalable human pose estimation using mmWave point cloud
Fast and scalable human pose estimation using mmWave point cloud
复制标题
DOI：
10.1145/3489517.3530522
复制
发表时间：
2022-04
期刊：
影响因子：
--
通讯作者：
Sizhe An;U. Ogras
中科院分区：
文献类型：
--
作者：
Sizhe An;U. Ogras
毫米波（mmWave）雷达能够以低成本和低计算要求实现高分辨率的人体姿态估计。然而，毫米波数据点云，处理算法的主要输入，是高度稀疏的，携带的信息比其他替代品，如视频帧少得多。此外，稀缺的标记毫米波数据阻碍了机器学习（ML）模型的开发，这些模型可以推广到看不见的场景。我们提出了一个快速和可扩展的人体姿态估计（FUSE）框架，结合多帧表示和元学习来解决这些挑战。实验评估表明，FUSE适应看不见的场景比目前的监督学习方法快4倍，估计人体关节坐标的平均绝对误差约为7 cm。
Millimeter-Wave (mmWave) radar can enable high-resolution human pose estimation with low cost and computational requirements. However, mmWave data point cloud, the primary input to processing algorithms, is highly sparse and carries significantly less information than other alternatives such as video frames. Furthermore, the scarce labeled mmWave data impedes the development of machine learning (ML) models that can generalize to unseen scenarios. We propose a fast and scalable human pose estimation (FUSE) framework that combines multi-frame representation and meta-learning to address these challenges. Experimental evaluations show that FUSE adapts to the unseen scenarios 4× faster than current supervised learning approaches and estimates human joint coordinates with about 7 cm mean absolute error.