TY - JOUR
T1 - Global relationship awareness 3-dimensional object detection using 4-dimensional radar
AU - Duan, Pianzhang
AU - Wang, Li
AU - Fang, Cheng
AU - Song, Ziying
AU - Gao, Ming
AU - Zhou, Mo
AU - Li, Ying
AU - Zhang, Yibo
AU - Fan, Wei
AU - Xu, Bin
N1 - Publisher Copyright:
© 2025 Elsevier Ltd.
PY - 2026/1/15
Y1 - 2026/1/15
N2 - 4D (4-dimensional) radar sensing technology is essential for high-precision autonomous driving perception systems, as its superior detection capabilities at increased distances, compared to traditional LiDAR (Light Detection and Ranging). However, due to the sparsity of point clouds and the low resolution of millimeter-wave radar, voxel-based methods may fail to detect distant or closely adjacent objects, leading to inadequate detection accuracy. To mitigate the accuracy issues arising from the sparse nature of point clouds in such scenarios, we propose a novel object detection network: GRA-Net (Global Relation-Aware object detection Network). By leveraging a self-attention mechanism, GRA-Net effectively learns critical features from each radar pillar, enhancing the network’s capacity to capture relevant information about nearby objects. Furthermore, we introduce a global perception module that integrates key features within the pillars and global features, mitigating the impact of point cloud sparsity, particularly in distant regions. We conducted a series of experiments to evaluate the performance of GRA-Net. On the Astyx HiRes 2019 dataset, our method achieved 33.63 mAP (mean Average Precision) and 43.93 mAP at the moderate level; On the View-of-Delft dataset, our method achieved 47.74 mAP in the entire annotated area and 69.25 mAP in the driving corridor area.
AB - 4D (4-dimensional) radar sensing technology is essential for high-precision autonomous driving perception systems, as its superior detection capabilities at increased distances, compared to traditional LiDAR (Light Detection and Ranging). However, due to the sparsity of point clouds and the low resolution of millimeter-wave radar, voxel-based methods may fail to detect distant or closely adjacent objects, leading to inadequate detection accuracy. To mitigate the accuracy issues arising from the sparse nature of point clouds in such scenarios, we propose a novel object detection network: GRA-Net (Global Relation-Aware object detection Network). By leveraging a self-attention mechanism, GRA-Net effectively learns critical features from each radar pillar, enhancing the network’s capacity to capture relevant information about nearby objects. Furthermore, we introduce a global perception module that integrates key features within the pillars and global features, mitigating the impact of point cloud sparsity, particularly in distant regions. We conducted a series of experiments to evaluate the performance of GRA-Net. On the Astyx HiRes 2019 dataset, our method achieved 33.63 mAP (mean Average Precision) and 43.93 mAP at the moderate level; On the View-of-Delft dataset, our method achieved 47.74 mAP in the entire annotated area and 69.25 mAP in the driving corridor area.
KW - 3-dimensional object detection
KW - 4-dimensional radar
KW - Autonomous driving
KW - Self-attention mechanism
UR - https://www.scopus.com/pages/publications/105023433943
U2 - 10.1016/j.engappai.2025.113318
DO - 10.1016/j.engappai.2025.113318
M3 - Article
AN - SCOPUS:105023433943
SN - 0952-1976
VL - 164
JO - Engineering Applications of Artificial Intelligence
JF - Engineering Applications of Artificial Intelligence
M1 - 113318
ER -