Robust 3D Object Detection from LiDAR-Radar Point Clouds via Cross-Modal Feature Augmentation
Jianning Deng,Gabriel Chan,Hantao Zhong,Chris Xiaoxuan Lu,Jianning Deng,Gabriel Chan,Hantao Zhong,Chris Xiaoxuan Lu
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 ...