LiRaFusion: Deep Adaptive LiDAR-Radar Fusion for 3D Object Detection

arXiv:2402.11735v1 Announce Type: cross
Abstract: We propose LiRaFusion to tackle LiDAR-radar fusion for 3D object detection to fill the performance gap of existing LiDAR-radar detectors. To improve the feature extraction capabilities from these two modalities, we design an early fusion module for joint voxel feature encoding, and a middle fusion module to adaptively fuse feature maps via a gated network. We perform extensive evaluation on nuScenes to demonstrate that LiRaFusion leverages the complementary information of LiDAR and radar effectively and …

3d object 3d object detection abstract arxiv capabilities cs.cv cs.ro design detection encoding extraction feature feature extraction fusion gap lidar performance radar type voxel