Diffusion Model is a Good Pose Estimator from 3D RF-Vision

arXiv:2403.16198v1 Announce Type: new
Abstract: Human pose estimation (HPE) from Radio Frequency vision (RF-vision) performs human sensing using RF signals that penetrate obstacles without revealing privacy (e.g., facial information). Recently, mmWave radar has emerged as a promising RF-vision sensor, providing radar point clouds by processing RF signals. However, the mmWave radar has a limited resolution with severe noise, leading to inaccurate and inconsistent human pose estimation. This work proposes mmDiff, a novel diffusion-based pose estimator tailored for noisy radar data. …

abstract arxiv cs.cv diffusion diffusion model estimator good however hpe human information obstacles privacy processing radar radio sensing sensor type vision