Weakly Supervised 3D Object Detection via Multi-Level Visual Guidance

arXiv:2312.07530v2 Announce Type: replace
Abstract: Weakly supervised 3D object detection aims to learn a 3D detector with lower annotation cost, e.g., 2D labels. Unlike prior work which still relies on few accurate 3D annotations, we propose a framework to study how to leverage constraints between 2D and 3D domains without requiring any 3D labels. Specifically, we employ visual data from three perspectives to establish connections between 2D and 3D domains. First, we design a feature-level constraint to align LiDAR and …

3d object 3d object detection arxiv cs.cv detection guidance object type via visual