Dynamic Edge Weights in Graph Neural Networks for 3D Object Detection

TL;DR

This paper introduces a novel graph neural network approach with dynamic edge weights and attention mechanisms for improved 3D object detection in LiDAR scans, emphasizing geometric feature retention and computational efficiency.

Contribution

It proposes an attention-based feature aggregation method with distance-aware down-sampling and masked attention in GNNs for enhanced 3D detection accuracy.

Findings

Achieves comparable results on KITTI dataset

Improves detection accuracy without high computational costs

Retains maximum geometric features of distant objects

Abstract

A robust and accurate 3D detection system is an integral part of autonomous vehicles. Traditionally, a majority of 3D object detection algorithms focus on processing 3D point clouds using voxel grids or bird's eye view (BEV). Recent works, however, demonstrate the utilization of the graph neural network (GNN) as a promising approach to 3D object detection. In this work, we propose an attention based feature aggregation technique in GNN for detecting objects in LiDAR scan. We first employ a distance-aware down-sampling scheme that not only enhances the algorithmic performance but also retains maximum geometric features of objects even if they lie far from the sensor. In each layer of the GNN, apart from the linear transformation which maps the per node input features to the corresponding higher level features, a per node masked attention by specifying different weights to different nodes in its first ring neighborhood is also performed. The masked attention implicitly accounts for the underlying neighborhood graph structure of every node and also eliminates the need of costly matrix operations thereby improving the detection accuracy without compromising the performance. The experiments on KITTI dataset show that our method yields comparable results for 3D object detection.