Saved in:
Bibliographic Details
Main Authors: Hao, Yiduo, Madani, Sohrab, Guan, Junfeng, Alloulah, Mohammed, Gupta, Saurabh, Hassanieh, Haitham
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2312.04519
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916211031277568
author Hao, Yiduo
Madani, Sohrab
Guan, Junfeng
Alloulah, Mohammed
Gupta, Saurabh
Hassanieh, Haitham
author_facet Hao, Yiduo
Madani, Sohrab
Guan, Junfeng
Alloulah, Mohammed
Gupta, Saurabh
Hassanieh, Haitham
contents The perception of autonomous vehicles using radars has attracted increased research interest due its ability to operate in fog and bad weather. However, training radar models is hindered by the cost and difficulty of annotating large-scale radar data. To overcome this bottleneck, we propose a self-supervised learning framework to leverage the large amount of unlabeled radar data to pre-train radar-only embeddings for self-driving perception tasks. The proposed method combines radar-to-radar and radar-to-vision contrastive losses to learn a general representation from unlabeled radar heatmaps paired with their corresponding camera images. When used for downstream object detection, we demonstrate that the proposed self-supervision framework can improve the accuracy of state-of-the-art supervised baselines by $5.8\%$ in mAP. Code is available at \url{https://github.com/yiduohao/Radical}.
format Preprint
id arxiv_https___arxiv_org_abs_2312_04519
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Bootstrapping Autonomous Driving Radars with Self-Supervised Learning
Hao, Yiduo
Madani, Sohrab
Guan, Junfeng
Alloulah, Mohammed
Gupta, Saurabh
Hassanieh, Haitham
Computer Vision and Pattern Recognition
The perception of autonomous vehicles using radars has attracted increased research interest due its ability to operate in fog and bad weather. However, training radar models is hindered by the cost and difficulty of annotating large-scale radar data. To overcome this bottleneck, we propose a self-supervised learning framework to leverage the large amount of unlabeled radar data to pre-train radar-only embeddings for self-driving perception tasks. The proposed method combines radar-to-radar and radar-to-vision contrastive losses to learn a general representation from unlabeled radar heatmaps paired with their corresponding camera images. When used for downstream object detection, we demonstrate that the proposed self-supervision framework can improve the accuracy of state-of-the-art supervised baselines by $5.8\%$ in mAP. Code is available at \url{https://github.com/yiduohao/Radical}.
title Bootstrapping Autonomous Driving Radars with Self-Supervised Learning
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2312.04519