RT-Pose: A 4D Radar Tensor-based 3D Human Pose Estimation and Localization Benchmark

TL;DR

This paper introduces RT-Pose, a novel radar-based 3D human pose estimation benchmark using 4D radar tensors, along with a new dataset and a single-stage architecture that outperforms previous methods in privacy-preserving human localization and pose estimation.

Contribution

The paper presents the RT-Pose dataset, a new benchmark with 4D radar tensors, and proposes HRRadarPose, the first single-stage model for radar-based 3D human pose estimation.

Findings

HRRadarPose achieves MPJPE of 9.91cm on RT-Pose.

RT-Pose dataset includes 72k frames with multi-modal data.

Radar-based HPE outperforms prior methods in privacy-preserving scenarios.

Abstract

Traditional methods for human localization and pose estimation (HPE), which mainly rely on RGB images as an input modality, confront substantial limitations in real-world applications due to privacy concerns. In contrast, radar-based HPE methods emerge as a promising alternative, characterized by distinctive attributes such as through-wall recognition and privacy-preserving, rendering the method more conducive to practical deployments. This paper presents a Radar Tensor-based human pose (RT-Pose) dataset and an open-source benchmarking framework. The RT-Pose dataset comprises 4D radar tensors, LiDAR point clouds, and RGB images, and is collected for a total of 72k frames across 240 sequences with six different complexity-level actions. The 4D radar tensor provides raw spatio-temporal information, differentiating it from other radar point cloud-based datasets. We develop an annotation process using RGB images and LiDAR point clouds to accurately label 3D human skeletons. In addition, we propose HRRadarPose, the first single-stage architecture that extracts the high-resolution representation of 4D radar tensors in 3D space to aid human keypoint estimation. HRRadarPose outperforms previous radar-based HPE work on the RT-Pose benchmark. The overall HRRadarPose performance on the RT-Pose dataset, as reflected in a mean per joint position error (MPJPE) of 9.91cm, indicates the persistent challenges in achieving accurate HPE in complex real-world scenarios. RT-Pose is available at https://huggingface.co/datasets/uwipl/RT-Pose.