In recent years graph neural network (GNN)-based approaches have become a popular strategy for processing point cloud data, regularly achieving state-of-the-art performance on a variety of tasks. To date, the research community has primarily focused on improving model expressiveness, with secondary thought given to how to design models that can run efficiently on resource constrained mobile devices including smartphones or mixed reality headsets. In this work we make a step towards improving the efficiency of these models by making the observation that these GNN models are heavily limited by the representational power of their first, feature extracting, layer. We find that it is possible to radically simplify these models so long as the feature extraction layer is retained with minimal degradation to model performance; further, we discover that it is possible to improve performance overall on ModelNet40 and S3DIS by improving the design of the feature extractor. Our approach reduces memory consumption by 20$\times$ and latency by up to 9.9$\times$ for graph layers in models such as DGCNN; overall, we achieve speed-ups of up to 4.5$\times$ and peak memory reductions of 72.5%.
CS-Net:Contribution-based Sampling Network for Point Cloud Simplification
School of Control Science and Engineering, Shandong University（山东大学控制科学与工程学院）；School of Software, Shandong University（山东大学软件学院）；School of Engineering and Sustainable Development, De Montfort University（德蒙福特大学工程与可持续发展学院）；Department of Computer Science, City University of Hong Kong（香港城市大学计算机科学系）
HPLFlowNet: Hierarchical Permutohedral Lattice FlowNet for Scene Flow Estimation on Large-scale Point Clouds
Stanford University（斯坦福大学）；University of California, Davis（加州大学戴维斯分校）；TuSimple（图森未来）
我们提出了一种新的深度神经网络架构，用于直接对大规模3D点云进行端到端场景流估计。受双边卷积层（BCL）的启发，我们提出了新的DownBCL，UpBCL和CorrBCL操作，从非结构化点云恢复结构信息，并融合来自两个连续点云的信息。操作离散和稀疏的permutohedral格点，我们的架构设计是节俭的计算成本。我们的模型可以有效地处理一对点云帧，每帧最多86K个点。我们的方法在FlyingThings3D和KITTI Scene Flow 2015数据集上实现了最先进的性能。此外，在合成数据上训练，我们的方法对真实世界的数据和不同的点密度表现出很强的泛化能力，而无需微调。
Towards Efficient Graph Convolutional Networks for Point Cloud Handling
Computer Vision Lab, ETH Zürich（苏黎世联邦理工学院计算机视觉实验室）；ETH Zürich（苏黎世联邦理工学院）；Johns Hopkins University（约翰斯·霍普金斯大学）；Microsoft（微软）；National University of Singapore（新加坡国立大学）；KU Leuven（鲁汶大学）
Efficient Weingarten Map and Curvature Estimation on Manifolds
School of Mathematics and Statistics, Beijing Institute of Technology（北京理工大学数学与统计学院）；Department of Computer Science, Princeton University（普林斯顿大学计算机科学系）；Department of Biostatistics, University of California, Los Angeles（加利福尼亚大学洛杉矶分校生物统计学系）；Beijing Key Laboratory on MCAACI（MCAACI北京市重点实验室）；Department of Mathematics, University of Wisconsin Madison（威斯康星大学麦迪逊分校数学系）
Enhancing Local Geometry Learning for 3D Point Cloud via Decoupling Convolution
Tokyo Institute of Technology（东京工业大学）；Artificial Intelligence Research Center, AIST（产业技术综合研究所人工智能研究中心）；Dalian University of Technology（大连理工大学）；PASCO Corporation（PASCO公司）
Less is More: Efficient Point Cloud Reconstruction via Multi-Head Decoders
School of Computing and Artificial Intelligence, Southwest Jiaotong University（计算机与人工智能学院，西南交通大学）；Engineering Research Center of Sustainable Urban Intelligent Transportation Ministry of Education, Chengdu, China（可持续城市智能交通工程研究中心，教育部，成都，中国）
Fast Attention-Based Simplification of LiDAR Point Clouds for Object Detection and Classification
Institute for Computer Science and Control (SZTAKI)（计算机科学与控制研究所）；School of Information Engineering（信息工程学院）；Faculty of Technology, Arts, and Culture（技术、艺术与文化学院）；School of Control Science and Engineering（控制科学与工程学院）
DensePoint: Learning Densely Contextual Representation for Efficient Point Cloud Processing
Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院）；Department of Automation, Tsinghua University（清华大学自动化系）
