SparseFusion: Efficient Sparse Multi-Modal Fusion Framework for Long-Range 3D Perception

arXiv:2403.10036v1 Announce Type: new
Abstract: Multi-modal 3D object detection has exhibited significant progress in recent years. However, most existing methods can hardly scale to long-range scenarios due to their reliance on dense 3D features, which substantially escalate computational demands and memory usage. In this paper, we introduce SparseFusion, a novel multi-modal fusion framework fully built upon sparse 3D features to facilitate efficient long-range perception. The core of our method is the Sparse View Transformer module, which selectively lifts regions of …

3d object 3d object detection abstract arxiv computational cs.cv detection features framework fusion however memory modal multi-modal novel object paper perception progress reliance scale type usage