Why Learn What Physics Already Knows? Realizing Agile mmWave-based Human Pose Estimation via Physics-Guided Preprocessing

2026-03-09

AI Summary

This work addresses the limitations of existing millimeter-wave human pose estimation methods, which rely heavily on data-driven preprocessing while neglecting known physical structures, resulting in model redundancy and constrained accuracy. To overcome this, the authors propose a physics-guided lightweight framework that explicitly models the intrinsic physical relationships among range, angle, and Doppler dimensions. By integrating human kinematic priors and employing multi-scale fusion that preserves spatial structure, maintains motion continuity, and aligns with anatomical constraints, the approach avoids redundant learning of well-established physical laws. Remarkably, using only a lightweight MLP regressor, the method achieves comparable accuracy while reducing model parameters by 55.7%–88.9%, enabling real-time deployment on resource-constrained platforms such as the Raspberry Pi.

Technology Category

• Computer Vision: Low Level & Physics-based Vision
• Intelligent Robots: State Estimation
• Machine Learning: Hardware-aware ML

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

We revisit millimeter-wave (mmWave) human pose estimation (HPE) from a signal preprocessing perspective. A single mmWave frame provides structured dimensions that map directly to human geometry and motion: range, angle, and Doppler, offering pose-aligned cues that are not explicitly present in RGB images. However, recent mmWave-based HPE systems require more parameters and compute resources yet yield lower estimation accuracy than vision baselines. We attribute this to preprocessing modules: most systems rely on data-driven modules to estimate phenomena that are already well-defined by mmWave sensing physics, whereas human pose could be captured more efficiently with explicit physical priors. To this end, we introduce processing modules that explicitly model mmWave's inter-dimensional correlations and human kinematics. Our design (1) couples range and angle to preserve spatial human structure, (2) leverages Doppler to retain human motion continuity, and (3) applies multi-scale fusion aligned with the human body. A lightweight MLP is involved as the regressor. In experiments, this framework reduces the number of parameters by 55.7-88.9% on the HPE task relative to existing mmWave baselines while maintaining competitive accuracy. Meanwhile, its lightweight nature enables real-time Raspberry Pi deployment. Code and deployment artifacts will be released upon acceptance.

Problem

• mmWave
• human pose estimation
• signal preprocessing
• physics priors
• real-time deployment

Innovation

• physics-guided preprocessing
• mmWave human pose estimation
• inter-dimensional correlation
• lightweight architecture
• real-time deployment