mm-Wave Radar Hand Shape Classification Using Deformable Transformers

TL;DR

This paper introduces a real-time mm-Wave radar-based hand shape classification method using deformable transformers, converting raw radar data into 3D point clouds, outperforming prior 2D image-based approaches.

Contribution

It presents a novel 3D radar neural network with deformable transformers for hand shape classification, surpassing previous methods that relied on 2D Range-Doppler images.

Findings

Significantly improved classification accuracy over prior methods.

Effective use of 3D sparse point clouds from raw radar data.

Real-time implementation on off-the-shelf radar sensor.

Abstract

A novel, real-time, mm-Wave radar-based static hand shape classification algorithm and implementation are proposed. The method finds several applications in low cost and privacy sensitive touchless control technology using 60 Ghz radar as the sensor input. As opposed to prior Range-Doppler image based 2D classification solutions, our method converts raw radar data to 3D sparse cartesian point clouds.The demonstrated 3D radar neural network model using deformable transformers significantly surpasses the performance results set by prior methods which either utilize custom signal processing or apply generic convolutional techniques on Range-Doppler FFT images. Experiments are performed on an internally collected dataset using an off-the-shelf radar sensor.