Abstract:Recent research has shown that mmWave radar sensing is effective for object detection in low visibility environments, which makes it an ideal technique in autonomous navigation systems such as autonomous vehicles. However, due to the characteristics of radar signals such as sparsity, low resolution, specularity, and high noise, it is still quite challenging to reconstruct 3D object shapes via mmWave radar sensing. Built on our recent proposed 3DRIMR (3D Reconstruction and Imaging via mmWave Radar), we introduce in this paper DeepPoint, a deep learning model that generates 3D objects in point cloud format that significantly outperforms the original 3DRIMR design. The model adopts a conditional Generative Adversarial Network (GAN) based deep neural network architecture. It takes as input the 2D depth images of an object generated by 3DRIMR's Stage 1, and outputs smooth and dense 3D point clouds of the object. The model consists of a novel generator network that utilizes a sequence of DeepPoint blocks or layers to extract essential features of the union of multiple rough and sparse input point clouds of an object when observed from various viewpoints, given that those input point clouds may contain many incorrect points due to the imperfect generation process of 3DRIMR's Stage 1. The design of DeepPoint adopts a deep structure to capture the global features of input point clouds, and it relies on an optimally chosen number of DeepPoint blocks and skip connections to achieve performance improvement over the original 3DRIMR design. Our experiments have demonstrated that this model significantly outperforms the original 3DRIMR and other standard techniques in reconstructing 3D objects.

A High Precision Reconstruction Method of Point Cloud for Three-dimensional Radar Imaging

A New Method of 3D Reconstruction Using the Point Cloud and Distance Images of Laser Radar

O-Unet: an Octree-Based Convolutional Neural Network for 3-D Radar Point Clouds Reconstruction

An 3D Point Cloud Reconstruction Approach from Single Image for Smart Power Plant

A novel rapid point-cloud surface reconstruction algorithm for laser imaging radar

DeepPoint: A Deep Learning Model for 3D Reconstruction in Point Clouds via mmWave Radar

A Point Cloud Improvement Method for High-Resolution 4D Mmwave Radar Imagery

3DRIMR: 3D Reconstruction and Imaging via mmWave Radar based on Deep Learning

PCGNet: Point Cloud Generation Network for 3D Perception using Monocular Images and Radar

4D High-Resolution Imagery of Point Clouds for Automotive Mmwave Radar

A 3-D Imaging Method Based on Radar Sensor Network Distributed along Circular Geometry

DREAM-PCD: Deep Reconstruction and Enhancement of mmWave Radar Pointcloud

Three-dimensional point cloud reconstruction of inverse synthetic aperture radar image sequences based on back projection and iterative closest point fusion

A Research on 3-D Point Clouds Reconstruction of SAR Images Driven by Electromagnetic Scattering Mechanisms

A 3d Imaging Technique for Circular Radar Sensor Networks Based on Radon Transform

R2P: A Deep Learning Model from mmWave Radar to Point Cloud

Scanning Radar Target Reconstruction Using Deep Convolutional Neural Network

3D Reconstruction of Multiple Objects by mmWave Radar on UAV

A Point Clouds Framework for 3-D Reconstruction of SAR Images Based on 3-D Parametric Electromagnetic Part Model.

A Point Cloud Enhancement Method for 4D mmWave Radar Imagery

High-resolution Automotive Radar Point Cloud Imaging and Processing