Point-GNN: Graph Neural Network for 3D Object Detection in a Point Cloud

Abstract

In this paper, we propose a graph neural network to detect objects from a LiDAR point cloud. Towards this end, we encode the point cloud efficiently in a fixed radius near-neighbors graph. We design a graph neural network, named Point-GNN, to predict the category and shape of the object that each vertex in the graph belongs to. In Point-GNN, we propose an auto-registration mechanism to reduce translation variance, and also design a box merging and scoring operation to combine detections from multiple vertices accurately. Our experiments on the KITTI benchmark show the proposed approach achieves leading accuracy using the point cloud alone and can even surpass fusion-based algorithms. Our results demonstrate the potential of using the graph neural network as a new approach for 3D object detection. The code is available https://github.com/WeijingShi/Point-GNN.

References in corpus (2)

• PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space
• Deep Continuous Fusion for Multi-Sensor 3D Object Detection

Cited by in corpus (27)

• A Review of Uncertainty Quantification in Deep Learning: Techniques, Applications and Challenges
• RangeRCNN: Towards Fast and Accurate 3D Object Detection with Range Image Representation
• PV-RCNN++: Point-Voxel Feature Set Abstraction With Local Vector Representation for 3D Object Detection
• Deep Learning for 3D Point Cloud Understanding: A Survey
• Learning Distilled Collaboration Graph for Multi-Agent Perception
• SIENet: Spatial Information Enhancement Network for 3D Object Detection from Point Cloud
• RadarNet: Exploiting Radar for Robust Perception of Dynamic Objects
• OCM3D: Object-Centric Monocular 3D Object Detection
• Pyramid R-CNN: Towards Better Performance and Adaptability for 3D Object Detection
• 3D Object Detection with Pointformer
• Deep Learning for Weakly-Supervised Object Detection and Object Localization: A Survey
• Graph Signal Processing for Geometric Data and Beyond: Theory and Applications
• An Anomaly Event Detection Method Based on GNN Algorithm for Multi-data Sources
• Lidar Point Cloud Guided Monocular 3D Object Detection
• Recognizing Predictive Substructures with Subgraph Information Bottleneck
• From Voxel to Point: IoU-guided 3D Object Detection for Point Cloud with Voxel-to-Point Decoder
• Exploiting Edge-Oriented Reasoning for 3D Point-based Scene Graph Analysis
• Anchor-free 3D Single Stage Detector with Mask-Guided Attention for Point Cloud
• Stereo RGB and Deeper LIDAR Based Network for 3D Object Detection
• EagerMOT: 3D Multi-Object Tracking via Sensor Fusion
• M3DSSD: Monocular 3D Single Stage Object Detector
• MBDF-Net: Multi-Branch Deep Fusion Network for 3D Object Detection
• Instance Segmentation GNNs for One-Shot Conformal Tracking at the LHC
• Branchy-GNN: a Device-Edge Co-Inference Framework for Efficient Point Cloud Processing
• Deeper or Wider Networks of Point Clouds with Self-attention?
• Exploring intermediate representation for monocular vehicle pose estimation
• Cascade Image Matting with Deformable Graph Refinement