Learning of 3D Graph Convolution Networks for Point Cloud Analysis

Dual-Graph Attention Convolution Network for 3-D Point Cloud Classification

Three-dimensional point cloud classification is fundamental but still challenging in 3-D vision. Existing graph-based deep learning methods fail to learn both low-level extrinsic and high-level intrinsic features together. These two levels of features are critical to improving classification accuracy. To this end, we propose a dual-graph attention convolution network (DGACN). The idea of DGACN is to use two types of graph attention convolution operations with a feedback graph feature fusion mechanism. Specifically, we exploit graph geometric attention convolution to capture low-level extrinsic features in 3-D space. Furthermore, we apply graph embedding attention convolution to learn multiscale low-level extrinsic and high-level intrinsic fused graph features together. Moreover, the points belonging to different parts in real-world 3-D point cloud objects are distinguished, which results in more robust performance for 3-D point cloud classification tasks than other competitive methods, in practice. Our extensive experimental results show that the proposed network achieves state-of-the-art performance on both the synthetic ModelNet40 and real-world ScanObjectNN datasets.

LSPConv: local spatial projection convolution for point cloud analysis

This study introduces a novel approach, Local Spatial Projection Convolution (LSPConv), for point cloud classification and semantic segmentation. Unlike conventional methods utilizing relative coordinates for local geometric information, our motivation stems from the inadequacy of existing techniques for representing the intricate spatial organization of unconsolidated and irregular 3D point clouds. To address this limitation, we propose a Local Spatial Projection Module utilizing a vector projection strategy, designed to capture comprehensive local spatial information more effectively. Moreover, recent studies emphasize the importance of anisotropic kernels for point cloud feature extraction, considering the distinct contributions of individual neighboring points. To cater to this requirement, we introduce the Feature Weight Assignment (FWA) Module to assign weights to neighboring points, enhancing the anisotropy crucial for accurate feature extraction. Additionally, we introduce an Anisotropic Relative Feature Encoding Module that adaptively encodes points based on their relative features, further amplifying the anisotropic characteristics. Our approaches achieve remarkable results for point cloud classification and segmentation in several benchmark datasets based on extensive qualitative and quantitative evaluation.

Long and Short-Range Dependency Graph Structure Learning Framework on Point Cloud

Graph convolutional neural networks can effectively process geometric data and thus have been successfully used in point cloud data representation. However, existing graph-based methods usually adopt the K-nearest neighbor (KNN) algorithm to construct graphs, which may not be optimal for point cloud analysis tasks, owning to the solution of KNN is independent of network training. In this paper, we propose a novel graph structure learning convolutional neural network (GSLCN) for multiple point cloud analysis tasks. The fundamental concept is to propose a general graph structure learning architecture (GSL) that builds long-range and short-range dependency graphs. To learn optimal graphs that best serve to extract local features and investigate global contextual information, respectively, we integrated the GSL with the designed graph convolution operator under a unified framework. Furthermore, we design the graph structure losses with some prior knowledge to guide graph learning during network training. The main benefit is that given labels and prior knowledge are taken into account in GSLCN, providing useful supervised information to build graphs and thus facilitating the graph convolution operation for the point cloud. Experimental results on challenging benchmarks demonstrate that the proposed framework achieves excellent performance for point cloud classification, part segmentation, and semantic segmentation.

Online Metro Origin-Destination Prediction via Heterogeneous Information Aggregation

Metro origin-destination prediction is a crucial yet challenging time-series analysis task in intelligent transportation systems, which aims to accurately forecast two specific types of cross-station ridership, i.e., Origin-Destination (OD) one and Destination-Origin (DO) one. However, complete OD matrices of previous time intervals can not be obtained immediately in online metro systems, and conventional methods only used limited information to forecast the future OD and DO ridership separately. In this work, we proposed a novel neural network module termed Heterogeneous Information Aggregation Machine (HIAM), which fully exploits heterogeneous information of historical data (e.g., incomplete OD matrices, unfinished order vectors, and DO matrices) to jointly learn the evolutionary patterns of OD and DO ridership. Specifically, an OD modeling branch estimates the potential destinations of unfinished orders explicitly to complement the information of incomplete OD matrices, while a DO modeling branch takes DO matrices as input to capture the spatial-temporal distribution of DO ridership. Moreover, a Dual Information Transformer is introduced to propagate the mutual information among OD features and DO features for modeling the OD-DO causality and correlation. Based on the proposed HIAM, we develop a unified Seq2Seq network to forecast the future OD and DO ridership simultaneously. Extensive experiments conducted on two large-scale benchmarks demonstrate the effectiveness of our method for online metro origin-destination prediction. Our code is resealed at https://github.com/HCPLab-SYSU/HIAM.

MSIDA-Net: Point Cloud Semantic Segmentation via Multi-Spatial Information and Dual Adaptive Blocks

Large-scale 3D point clouds are rich in geometric shape and scale information but they are also scattered, disordered and unevenly distributed. These characteristics lead to difficulties in learning point cloud semantic segmentations. Although many works have performed well in this task, most of them lack research on spatial information, which limits the ability to learn and understand the complex geometric structure of point cloud scenes. To this end, we propose the multispatial information and dual adaptive (MSIDA) module, which consists of a multispatial information encoding (MSI) block and dual adaptive (DA) blocks. The MSI block transforms the information of the relative position of each centre point and its neighbouring points into a cylindrical coordinate system and spherical coordinate system. Then the spatial information among the points can be re-represented and encoded. The DA blocks include a Coordinate System Attention Pooling Fusion (CSAPF) block and a Local Aggregated Feature Attention (LAFA) block. The CSAPF block weights and fuses the local features in the three coordinate systems to further learn local features, while the LAFA block weights the local aggregated features in the three coordinate systems to better understand the scene in the local region. To test the performance of the proposed method, we conducted experiments on the S3DIS, Semantic3D and SemanticKITTI datasets and compared the proposed method with other networks. The proposed method achieved 73%, 77.8% and 59.8% mean Intersection over Union (mIoU) on the S3DIS, Semantic3D and SemanticKITTI datasets, respectively.

3D Instance Segmentation and Object Detection Framework Based on the Fusion of Lidar Remote Sensing and Optical Image Sensing

Since single sensor and high-density point cloud data processing have certain direct processing limitations in urban traffic scenarios, this paper proposes a 3D instance segmentation and object detection framework for urban transportation scenes based on the fusion of Lidar remote sensing technology and optical image sensing technology. Firstly, multi-source and multi-mode data pre-fusion and alignment of Lidar and camera sensor data are effectively carried out, and then a unique and innovative network of stereo regional proposal selective search-driven DAGNN is constructed. Finally, using the multi-dimensional information interaction, three-dimensional point clouds with multi-features and unique concave-convex geometric characteristics are instance over-segmented and clustered by the hypervoxel storage in the remarkable octree and growing voxels. Finally, the positioning and semantic information of significant 3D object detection in this paper are visualized by multi-dimensional mapping of the boundary box. The experimental results validate the effectiveness of the proposed framework with excellent feedback for small objects, object stacking, and object occlusion. It can be a remediable or alternative plan to a single sensor and provide an essential theoretical and application basis for remote sensing, autonomous driving, environment modeling, autonomous navigation, and path planning under the V2X intelligent network space– ground integration in the future.