Range Conditioned Dilated Convolutions for Scale Invariant 3D Object Detection
This work addresses the problem of accurate 3D object detection for autonomous driving systems, particularly improving long-range performance, though it is incremental in advancing range-based methods.
The paper tackles scale sensitivity in 3D object detection from LiDAR range images by proposing a range-conditioned dilation layer and localized soft range gating, achieving new state-of-the-art performance on the Waymo Open Dataset with unparalleled long-range detection.
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 a 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.