Range Conditioned Dilated Convolutions for Scale Invariant 3D Object Detection

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 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.

arXiv commentCoRL 2020

References in corpus (7)

• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• Rethinking Atrous Convolution for Semantic Image Segmentation
• Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs)
• Layer Normalization
• Frustum ConvNet: Sliding Frustums to Aggregate Local Point-Wise Features for Amodal 3D Object Detection
• StarNet: Targeted Computation for Object Detection in Point Clouds
• From Points to Parts: 3D Object Detection from Point Cloud with Part-aware and Part-aggregation Network

Cited by in corpus (4)

• Center-based 3D Object Detection and Tracking
• RangeRCNN: Towards Fast and Accurate 3D Object Detection with Range Image Representation
• Pyramid R-CNN: Towards Better Performance and Adaptability for 3D Object Detection
• LiDAR R-CNN: An Efficient and Universal 3D Object Detector