Automatic Labelling & Semantic Segmentation with 4D Radar Tensors
This addresses the need for efficient perception in autonomous driving by providing an incremental improvement in radar-based segmentation.
The paper tackles the problem of automatic labeling and semantic segmentation for automotive datasets using 4D radar data, achieving over 65% of LiDAR detection performance, a 13.2% improvement in vehicle detection probability, and a 0.54 m reduction in Chamfer distance compared to literature variants.
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.