Accurate Detection Using Multi-Resolution Cascaded MIMO Radar


Sohrab Madani*
Junfeng Guan*
Waleed Ahmed*
Saurabh Gupta
Haitham Hassanieh

University of Illinois Urbana-Champaign
* indicates equal contribution

Abstract

Millimeter wave (mmWave) radars are becoming a more popular sensing modality in self-driving cars due to their favorable characteristics in adverse weather. Yet, they currently lack sufficient spatial resolution for semantic scene understanding. In this paper, we present Radatron, a system capable of accurate object detection using mmWave radar as a stand-alone sensor. To enable Radatron, we introduce a first-of-its-kind, high-resolution automotive radar dataset collected with a cascaded Multiple Input Multiple Output (MIMO) radar. Our radar achieves 5cm range resolution and 1.2 degree angular resolution, 10x finer than other publicly available datasets. We also develop a novel hybrid radar processing and deep learning approach to achieve high vehicle detection accuracy. We train and extensively evaluate Radatron to show it achieves 92.6% AP50 and 56.3% AP75 accuracy in 2D bounding box detection, an 8% and 15.9% improvement over prior art respectively.


Paper

Accurate Detection Using Multi-Resolution Cascaded MIMO Radar
Sohrab Madani*, Junfeng Guan*, Waleed Ahmed*, Saurabh Gupta, Haitham Hassanieh
European Conference on Computer Vision (ECCV), 2022
* indicates equal contribution

[Paper] [Supp]




Project Overview Video



Radatron Dataset

[Documentation] [Processed Dataset] [Raw Radar Data]


Radatron Network Code

[GitHub]


Radar Signal Processing Code

[GitHub]


Results

Examples from our test set. Ground truth marked in green and predictions in red. Top row shows the rriginal scene. Second row shows Radatron's performance overlaid on distortion compensated radar heatmaps. The bottom two rows show the performances of stand-alone cascaded and radar in prior work baselines along with their input heatmaps respectively.


University of Illinois Urbana-Champaign | SyNRG