Raw Radar data based Object Detection and Heading estimation using Cross Attention

TL;DR

This paper introduces a deep learning framework utilizing raw radar data with cross-attention mechanisms for improved object detection and heading estimation, enhancing accuracy and reducing complexity in autonomous driving perception.

Contribution

It presents a novel end-to-end neural network that combines refined dataset annotations, transformer-inspired cross-attention fusion, and center-offset maps for better radar-based perception.

Findings

Detection mAP improved by 5%

Model complexity reduced by 23%

Extended to include heading estimation

Abstract

Radar is an inevitable part of the perception sensor set for autonomous driving functions. It plays a gap-filling role to complement the shortcomings of other sensors in diverse scenarios and weather conditions. In this paper, we propose a Deep Neural Network (DNN) based end-to-end object detection and heading estimation framework using raw radar data. To this end, we approach the problem in both a Data-centric and model-centric manner. We refine the publicly available CARRADA dataset and introduce Bivariate norm annotations. Besides, the baseline model is improved by a transformer inspired cross-attention fusion and further center-offset maps are added to reduce localisation error. Our proposed model improves the detection mean Average Precision (mAP) by 5%, while reducing the model complexity by almost 23%. For comprehensive scene understanding purposes, we extend our model for heading estimation. The improved ground truth and proposed model is available at Github