Doppler Prompting for Stable mmWave-based Human Pose Estimation

TL;DR

This paper introduces PULSE, a novel method that uses Doppler signatures as confidence-aware prompts to improve the stability and accuracy of mmWave-based human pose estimation.

Contribution

PULSE is the first approach to convert Doppler signatures into motion prompts and integrate them with spatial data for more stable pose estimation.

Findings

PULSE improves pose accuracy across multiple datasets.

It enhances temporal stability of human pose trajectories.

The method effectively suppresses spurious Doppler cues.

Abstract

Millimeter-wave (mmWave) enables privacy-preserving, illumination-robust human pose estimation (HPE), with each mmWave frame represented as a range-angle-Doppler tensor, providing spatial magnitude for localization and Doppler signatures for motion-related cues. However, existing mmWave-based HPE methods either underutilize or na\"ively fuse Doppler signatures with spatial magnitude, disregarding their distinct physical semantics. As a result, non-human Doppler signatures can be misinterpreted as human motion cues, leading to jittery trajectories. We propose PULSE, which converts Doppler signatures into confidence-aware motion prompts and injects them into spatial magnitude reasoning through constrained interactions. By screening Doppler prompts before they influence prediction, PULSE first suppresses spurious spectral motion cues and then uses the screened prompts to stabilize prediction. Across three datasets spanning single- and multi-person settings, PULSE consistently improves pose accuracy and temporal stability, indicating that controlled Doppler prompting is a practical direction for stable mmWave HPE.