Abstract:Background: Due to the refinement of region of the interests (RoIs), two-stage 3D detection algorithms can usually obtain better performance compared with most single-stage detectors. However, most two-stage methods adopt feature connection, to aggregate the grid point features using multi-scale RoI pooling in the second stage. This connection mode does not consider the correlation between multi-scale grid features. Methods: In the first stage, we employ 3D sparse convolution and 2D convolution to fully extract rich semantic features. Then, a small number of coarse RoIs are predicted based region proposal network (RPN) on generated bird’s eye view (BEV) map. After that, we adopt voxel RoI-pooling strategy to aggregate the neighborhood nonempty voxel features of each grid point in RoI in the last two layers of 3D sparse convolution. In this way, we obtain two aggregated features from 3D sparse voxel space for each grid point. Next, we design an attention feature fusion module. This module includes a local and a global attention layer, which can fully integrate the grid point features from different voxel layers. Results: We carry out relevant experiments on the Karlsruhe Institute of Technology and Toyota Technological Institute (KITTI) dataset. The average precisions of our proposed method are 88.21%, 81.51%, 77.07% on three difficulty levels (easy, moderate, and hard, respectively) for 3D detection, and 92.30%, 90.19%, 86.00% on three difficulty levels (easy, moderate, and hard, respectively) for BEV detection. Conclusions: In this paper, we propose a novel two-stage 3D detection algorithm named Grid Attention Fusion Region-based Convolutional Neural Network (GAF-RCNN) from point cloud. Because we integrate multi-scale RoI grid features with attention mechanism in the refinement stage, different multi-scale features can be better correlated, achieving a competitive level compared with other well tested detection algorithms. This 3D object detection has important implications for robot and cobot technology.

three-dimensional grid data processing method and system based on a neural network

Study on three﹣dimensional effect correction of surface heat flux estimation based on neural network

A Generalized Spatial Autoregressive Neural Network Method for Three-Dimensional Spatial Interpolation

Research on the Visual Image Processing Based on Neural Network

A three-dimensional advancing front technique to generate grids based on the neural networks

Analyzing 3D Model Visual Features Based on Neural Networks

GridFormer: Point-Grid Transformer for Surface Reconstruction

MeshNet: Mesh Neural Network for 3D Shape Representation

A feature fusion-based attention graph convolutional network for 3D classification and segmentation

Three-dimensional object normal vector, geometry and material acquisition method based on neural network

Learned Gridification for Efficient Point Cloud Processing

An improved fused feature residual network for 3D point cloud data

GAF-RCNN: Grid Attention Fusion 3D Object Detection from Point Cloud

FatNet: A Feature-attentive Network for 3D Point Cloud Processing

Grounding and Enhancing Grid-based Models for Neural Fields

ArticleNovel intelligent spatiotemporal grid earthquake early-warning modelDaoye Zhu 1

A 3D Point Cloud Classification Method Based on Adaptive Graph Convolution and Global Attention

SDANet: spatial deep attention-based for point cloud classification and segmentation

3D model retrieval based on interactive attention CNN and multiple features

Novel Intelligent Spatiotemporal Grid Earthquake Early-Warning Model

Grid Network: Feature Extraction in Anisotropic Perspective for Hyperspectral Image Classification