Abstract:The rapid growth in technologies for 3D sensors has made point cloud data increasingly available in different applications such as autonomous driving, robotics, and virtual and augmented reality. This raises a growing need for deep learning methods to process the data. Point clouds are difficult to be used directly as inputs in several deep learning techniques. The difficulty is raised by the unstructured and unordered nature of the point cloud data. So, machine learning models built for images or videos cannot be used directly on point cloud data. Although the research in the field of point clouds has gained high attention and different methods have been developed over the decade, very few research works directly with point cloud data, and most of them convert the point cloud data into 2D images or voxels by performing some pre-processing that causes information loss. Methods that directly work on point clouds are in the early stage and this affects the performance and accuracy of the models. Advanced techniques in classical convolutional neural networks, such as the attention mechanism, need to be transferred to the methods directly working with point clouds. In this research, an attention mechanism is proposed to be added to deep convolutional neural networks that process point clouds directly. The attention module was proposed based on specific pooling operations which are designed to be applied directly to point clouds to extract vital information from the point clouds. Segmentation of the ShapeNet dataset was performed to evaluate the method. The mean intersection over union (mIoU) score of the proposed framework was increased after applying the attention method compared to a base state-of-the-art framework that does not have the attention mechanism.

Attention-Based Deep Neural Network for Point Cloud Learning

PointAttN: You Only Need Attention for Point Cloud Completion

PointCloud-At: Point Cloud Convolutional Neural Networks with Attention for 3D Data Processing

Point Cloud Understanding via Attention-Driven Contrastive Learning

Deep Neural Network for Point Sets Based on Local Feature Integration

DeepICP: An End-to-End Deep Neural Network for 3D Point Cloud Registration

Point Attention Network for Semantic Segmentation of 3D Point Clouds

Learning Representative Features by Deep Attention Network for 3D Point Cloud Registration

Adaptive deep learning-based neighborhood search method for point cloud

Point Attention Network for Point Cloud Semantic Segmentation.

Point Cloud Deep Learning Network Based on Local Domain Multi-Level Feature

BiPointNet: Binary Neural Network for Point Clouds

Point clouds learning with attention-based graph convolution networks

UFO-Net: A Linear Attention-Based Network for Point Cloud Classification

Deep Global Features for Point Cloud Alignment

AttPNet: Attention-Based Deep Neural Network for 3D Point Set Analysis

Global Attention-Guided Dual-Domain Point Cloud Feature Learning for Classification and Segmentation

Multi-Head Self-Attention for 3D Point Cloud Classification

CloudAttention: Efficient Multi-Scale Attention Scheme For 3D Point Cloud Learning

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

PACNet: A High-precision Point Cloud Registration Network Based on Deep Learning