Fast and Scalable Human Pose Estimation using mmWave Point Cloud

Abstract

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$\times$ faster than current supervised learning approaches and estimates human joint coordinates with about 7 cm mean absolute error.

arXiv commentAccepted for Design Automation Conference (DAC) 2022

References in corpus (5)

• Adam: A Method for Stochastic Optimization
• Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
• mm-Pose: Real-Time Human Skeletal Posture Estimation using mmWave Radars and CNNs
• Rethinking Motion Representation: Residual Frames with 3D ConvNets for Better Action Recognition
• Few-Shot Meta-Learning on Point Cloud for Semantic Segmentation

Cited by in corpus (2)

• mmGAT: Pose Estimation by Graph Attention with Mutual Features from mmWave Radar Point Cloud
• PCFEx: Point Cloud Feature Extraction for Graph Neural Networks