Paper Info
Title
A Hierarchical Graph Network for 3D Object Detection on Point Clouds
Abstract
3D object detection on point clouds finds many applications. However, most known point cloud object detection methods did not adequately accommodate the characteristics (e.g., sparsity) of point clouds, and thus some key semantic information (e.g., shape information) is not well captured. In this paper, we propose a new graph convolution (GConv) based hierarchical graph network (HGNet) for 3D object detection, which processes raw point clouds directly to predict 3D bounding boxes. HGNet effectively captures the relationship of the points and utilizes the multi-level semantics for object detection. Specially, we propose a novel shape-attentive GConv (SA-GConv) to capture the local shape features, by modelling the relative geometric positions of points to describe object shapes. An SA-GConv based U-shape network captures the multi-level features, which are mapped into an identical feature space by an improved voting module and then further utilized to generate proposals. Next, a new GConv based Proposal Reasoning Module reasons on the proposals considering the global scene semantics, and the bounding boxes are then predicted. Consequently, our new framework outperforms state-of-the-art methods on two large-scale point cloud datasets, by ~4% mean average precision (mAP) on SUN RGB-D and by ~3% mAP on ScanNet-V2.
Year
DOI
Venue
2020
10.1109/CVPR42600.2020.00047
2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Keywords
DocType
ISSN
known point cloud object detection,key semantic information,shape information,hierarchical graph network,raw point clouds,shape-attentive GConv,local shape features,object shapes,U-shape network,large-scale point cloud datasets
Conference
1063-6919
ISBN
Citations 
PageRank 
978-1-7281-7169-2
2
0.35
References 
Authors
28
6
Name
Order
Citations
PageRank
Jintai Chen144.12
Biwen Lei251.07
Qingyu Song341.42
Haochao Ying47310.03
Danny Z. Chen51713165.02
Jian Wu693395.62