CoRL 2020, Spotlight Talk 134: Range Conditioned Dilated Convolutions for Scale Invariant 3D Object Detection

CoRL 2020

Details
"**Range Conditioned Dilated Convolutions for Scale Invariant 3D Object Detection** Alex Bewley (Google)*; Pei Sun (Waymo); Thomas Mensink (Google Research / University of Amsterdam); Dragomir Anguelov (Waymo); Cristian Sminchisescu (Google) Publication: http://corlconf.github.io/paper_134/ **Abstract** This paper presents a novel 3D object detection framework that processes LiDAR data directly on its native representation: range images. Benefiting from the compactness of range images, 2D convolutions can efficiently process dense LiDAR data of the scene. To overcome scale sensitivity in this perspective view, a novel range-conditioned dilation (RCD) layer is proposed to dynamically adjust a continuous dilation rate as a function of the measured range. Furthermore, localized soft range gating combined with a 3D box-refinement stage improves robustness in occluded areas, and produces overall more accurate bounding box predictions. On the public large-scale Waymo Open Dataset, our method sets a new baseline for range-based 3D detection, outperforming multiview and voxel-based methods over all ranges with unparalleled performance at long range detection.

Comments
loading...