Abstract:Recently, graph-based and Transformer-based deep learning networks have demonstrated excellent performances on various point cloud tasks. Most of the existing graph methods are based on static graph, which take a fixed input to establish graph relations. Moreover, many graph methods apply maximization and averaging to aggregate neighboring features, so that only a single neighboring point affects the feature of centroid or different neighboring points have the same influence on the centroid's feature, which ignoring the correlation and difference between points. Most Transformer-based methods extract point cloud features based on global attention and lack the feature learning on local neighbors. To solve the problems of these two types of models, we propose a new feature extraction block named Graph Transformer and construct a 3D point point cloud learning network called GTNet to learn features of point clouds on local and global patterns. Graph Transformer integrates the advantages of graph-based and Transformer-based methods, and consists of Local Transformer and Global Transformer modules. Local Transformer uses a dynamic graph to calculate all neighboring point weights by intra-domain cross-attention with dynamically updated graph relations, so that every neighboring point could affect the features of centroid with different weights; Global Transformer enlarges the receptive field of Local Transformer by a global self-attention. In addition, to avoid the disappearance of the gradient caused by the increasing depth of network, we conduct residual connection for centroid features in GTNet; we also adopt the features of centroid and neighbors to generate the local geometric descriptors in Local Transformer to strengthen the local information learning capability of the model. Finally, we use GTNet for shape classification, part segmentation and semantic segmentation tasks in this paper.

Transformer-based Point Cloud Generation Network

Point Cloud Completion Cascade Optimization Network Based on Feature Fusion

PU-CTG: A Point Cloud Upsampling Network Using Transformer Fusion and GRU Correction

Point cloud upsampling via a coarse-to-fine network with transformer-encoder

3DCTN: 3D Convolution-Transformer Network for Point Cloud Classification

PU-Transformer: Point Cloud Upsampling Transformer

EGCT: Enhanced Graph Convolutional Transformer for 3D Point Cloud Representation Learning

Patch-Wise Point Cloud Generation: A Divide-and-Conquer Approach

Conversion of Point Cloud Data to 3D Models Using PointNet++ and Transformer

Point Cloud Upsampling via Cascaded Refinement Network

A Coarse-to-Fine Transformer-Based Network for 3D Reconstruction from Non-Overlapping Multi-View Images

Group-in-Group Relation-Based Transformer for 3D Point Cloud Learning

An 3D Point Cloud Reconstruction Approach from Single Image for Smart Power Plant

3DPCTN: Two 3D Local-Object Point-Cloud-Completion Transformer Networks Based on Self-Attention and Multi-Resolution

Stratified Transformer for 3D Point Cloud Segmentation

PCT: Point cloud transformer

HFF-Net: Hierarchical Feature Fusion Network for Point Cloud Generation with Point Transformers

Multistage Adaptive Point-Growth Network for Dense Point Cloud Completion

Progressive Point Cloud Deconvolution Generation Network

PointGPT: Auto-regressively Generative Pre-training from Point Clouds

GTNet: Graph Transformer Network for 3D Point Cloud Classification and Semantic Segmentation