mm-Pose: Real-Time Human Skeletal Posture Estimation using mmWave Radars and CNNs

TL;DR

This paper introduces mm-Pose, a real-time skeletal posture estimation system using mmWave radar and CNNs, capable of detecting over 15 joints, robust to lighting and weather, with applications in various fields.

Contribution

The paper presents the first method to detect more than 15 skeletal joints using mmWave radar signals and introduces a novel radar-to-image representation for improved accuracy.

Findings

Successfully detected and tracked human skeletal joints in real-time.

Validated the system with four primary human motions.

Achieved accurate 3D joint position predictions.

Abstract

In this paper, mm-Pose, a novel approach to detect and track human skeletons in real-time using an mmWave radar, is proposed. To the best of the authors' knowledge, this is the first method to detect >15 distinct skeletal joints using mmWave radar reflection signals. The proposed method would find several applications in traffic monitoring systems, autonomous vehicles, patient monitoring systems and defense forces to detect and track human skeleton for effective and preventive decision making in real-time. The use of radar makes the system operationally robust to scene lighting and adverse weather conditions. The reflected radar point cloud in range, azimuth and elevation are first resolved and projected in Range-Azimuth and Range-Elevation planes. A novel low-size high-resolution radar-to-image representation is also presented, that overcomes the sparsity in traditional point cloud data and offers significant reduction in the subsequent machine learning architecture. The RGB channels were assigned with the normalized values of range, elevation/azimuth and the power level of the reflection signals for each of the points. A forked CNN architecture was used to predict the real-world position of the skeletal joints in 3-D space, using the radar-to-image representation. The proposed method was tested for a single human scenario for four primary motions, (i) Walking, (ii) Swinging left arm, (iii) Swinging right arm, and (iv) Swinging both arms to validate accurate predictions for motion in range, azimuth and elevation. The detailed methodology, implementation, challenges, and validation results are presented.