PointRGCN: Graph Convolution Networks for 3D Vehicles Detection Refinement

TL;DR

PointRGCN introduces a graph convolutional network-based pipeline for 3D vehicle detection from LiDAR data, utilizing proposal and context aggregation to improve accuracy and achieve state-of-the-art results.

Contribution

The paper presents a novel graph-based 3D detection pipeline with residual and contextual GCNs for proposal refinement, advancing LiDAR-based vehicle detection.

Findings

Achieves state-of-the-art performance on bird's eye view detection

Utilizes residual GCNs for proposal classification and regression

Incorporates contextual GCNs for proposal refinement

Abstract

In autonomous driving pipelines, perception modules provide a visual understanding of the surrounding road scene. Among the perception tasks, vehicle detection is of paramount importance for a safe driving as it identifies the position of other agents sharing the road. In our work, we propose PointRGCN: a graph-based 3D object detection pipeline based on graph convolutional networks (GCNs) which operates exclusively on 3D LiDAR point clouds. To perform more accurate 3D object detection, we leverage a graph representation that performs proposal feature and context aggregation. We integrate residual GCNs in a two-stage 3D object detection pipeline, where 3D object proposals are refined using a novel graph representation. In particular, R-GCN is a residual GCN that classifies and regresses 3D proposals, and C-GCN is a contextual GCN that further refines proposals by sharing contextual information between multiple proposals. We integrate our refinement modules into a novel 3D detection pipeline, PointRGCN, and achieve state-of-the-art performance on the easy difficulty for the bird eye view detection task.