Robust 3D Object Detection from LiDAR-Radar Point Clouds via Cross-Modal Feature Augmentation

arXiv:2309.17336v2 Announce Type: replace
Abstract: This paper presents a novel framework for robust 3D object detection from point clouds via cross-modal hallucination. Our proposed approach is agnostic to either hallucination direction between LiDAR and 4D radar. We introduce multiple alignments on both spatial and feature levels to achieve simultaneous backbone refinement and hallucination generation. Specifically, spatial alignment is proposed to deal with the geometry discrepancy for better instance matching between LiDAR and radar. The feature alignment step further bridges the …

3d object 3d object detection abstract arxiv augmentation cs.cv cs.ro detection feature framework hallucination lidar modal multiple novel object paper radar robust spatial type via