Automatic Labelling & Semantic Segmentation with 4D Radar Tensors

TL;DR

This paper introduces an automatic labeling method for automotive datasets using LiDAR and camera data, and applies it to train a semantic segmentation network on 4D radar tensors, achieving promising results in vehicle detection and spatial accuracy.

Contribution

It presents a novel automatic labeling process for radar data and a semantic segmentation network that leverages 4D radar tensors for improved vehicle detection.

Findings

Achieved over 65% of LiDAR detection performance

Improved vehicle detection probability by 13.2%

Reduced Chamfer distance by 0.54 meters

Abstract

In this paper, an automatic labelling process is presented for automotive datasets, leveraging on complementary information from LiDAR and camera. The generated labels are then used as ground truth with the corresponding 4D radar data as inputs to a proposed semantic segmentation network, to associate a class label to each spatial voxel. Promising results are shown by applying both approaches to the publicly shared RaDelft dataset, with the proposed network achieving over 65% of the LiDAR detection performance, improving 13.2% in vehicle detection probability, and reducing 0.54 m in terms of Chamfer distance, compared to variants inspired from the literature.