PointSIFT: A SIFT-like Network Module for 3D Point Cloud Semantic Segmentation

Abstract

Recently, 3D understanding research sheds light on extracting features from point cloud directly, which requires effective shape pattern description of point clouds. Inspired by the outstanding 2D shape descriptor SIFT, we design a module called PointSIFT that encodes information of different orientations and is adaptive to scale of shape. Specifically, an orientation-encoding unit is designed to describe eight crucial orientations, and multi-scale representation is achieved by stacking several orientation-encoding units. PointSIFT module can be integrated into various PointNet-based architecture to improve the representation ability. Extensive experiments show our PointSIFT-based framework outperforms state-of-the-art method on standard benchmark datasets. The code and trained model will be published accompanied by this paper.

References in corpus (8)

• PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation
• Spectral Networks and Locally Connected Networks on Graphs
• PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space
• Fully Convolutional Networks for Semantic Segmentation
• ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes
• Multi-view Convolutional Neural Networks for 3D Shape Recognition
• FPNN: Field Probing Neural Networks for 3D Data
• SyncSpecCNN: Synchronized Spectral CNN for 3D Shape Segmentation

Cited by in corpus (56)

• Deep Learning for Image and Point Cloud Fusion in Autonomous Driving: A Review
• PointRCNN: 3D Object Proposal Generation and Detection from Point Cloud
• IPOD: Intensive Point-based Object Detector for Point Cloud
• Directionally Constrained Fully Convolutional Neural Network For Airborne Lidar Point Cloud Classification
• Relation-Shape Convolutional Neural Network for Point Cloud Analysis
• Adversarial Attack and Defense on Point Sets
• GA-NET: Global Attention Network for Point Cloud Semantic Segmentation
• STD: Sparse-to-Dense 3D Object Detector for Point Cloud
• Advancements in Point Cloud-Based 3D Defect Detection and Classification for Industrial Systems: A Comprehensive Survey
• Deep Learning for 3D Point Clouds: A Survey
• Deep Learning Based 3D Segmentation: A Survey
• 3D shape sensing and deep learning-based segmentation of strawberries
• VV-Net: Voxel VAE Net with Group Convolutions for Point Cloud Segmentation
• Patch-based Progressive 3D Point Set Upsampling
• MKConv: Multidimensional Feature Representation for Point Cloud Analysis
• A Survey of Simultaneous Localization and Mapping with an Envision in 6G Wireless Networks
• DensePoint: Learning Densely Contextual Representation for Efficient Point Cloud Processing
• Interpolated Convolutional Networks for 3D Point Cloud Understanding
• Morphing and Sampling Network for Dense Point Cloud Completion
• Rotation Invariant Point Cloud Classification: Where Local Geometry Meets Global Topology
• Pointwise Rotation-Invariant Network with Adaptive Sampling and 3D Spherical Voxel Convolution
• Multi-Modal 3D Object Detection in Autonomous Driving: a Survey
• GSIP: Green Semantic Segmentation of Large-Scale Indoor Point Clouds
• DAPnet: A Double Self-attention Convolutional Network for Point Cloud Semantic Labeling
• Deformable Filter Convolution for Point Cloud Reasoning
• SOE-Net: A Self-Attention and Orientation Encoding Network for Point Cloud based Place Recognition
• 3D-Rotation-Equivariant Quaternion Neural Networks
• Robust Multi-Modality Multi-Object Tracking
• Benchmarking Deep Learning Architectures for Urban Vegetation Point Cloud Semantic Segmentation from MLS
• SRINet: Learning Strictly Rotation-Invariant Representations for Point Cloud Classification and Segmentation
• ASAP-Net: Attention and Structure Aware Point Cloud Sequence Segmentation
• PointManifold: Using Manifold Learning for Point Cloud Classification
• DLA-Net: Learning Dual Local Attention Features for Semantic Segmentation of Large-Scale Building Facade Point Clouds
• PointAtrousGraph: Deep Hierarchical Encoder-Decoder with Point Atrous Convolution for Unorganized 3D Points
• MeshNet: Mesh Neural Network for 3D Shape Representation
• Multiview Based 3D Scene Understanding On Partial Point Sets
• Are We Hungry for 3D LiDAR Data for Semantic Segmentation? A Survey and Experimental Study
• REGNet: REgion-based Grasp Network for End-to-end Grasp Detection in Point Clouds
• 3DPVNet: Patch-level 3D Hough Voting Network for 6D Pose Estimation
• Airborne LiDAR Point Cloud Classification with Graph Attention Convolution Neural Network
• MANet: Multimodal Attention Network based Point- View fusion for 3D Shape Recognition
• An Evaluation of RGB and LiDAR Fusion for Semantic Segmentation
• Neural Style Transfer for Point Clouds
• Gap Completion in Point Cloud Scene occluded by Vehicles using SGC-Net
• Investigate Indistinguishable Points in Semantic Segmentation of 3D Point Cloud
• AdvectiveNet: An Eulerian-Lagrangian Fluidic reservoir for Point Cloud Processing
• CT-block: a novel local and global features extractor for point cloud
• LGENet: Local and Global Encoder Network for Semantic Segmentation of Airborne Laser Scanning Point Clouds
• MG-SAGC: A multiscale graph and its self-adaptive graph convolution network for 3D point clouds
• GraNet: Global Relation-aware Attentional Network for ALS Point Cloud Classification
• MNEW: Multi-domain Neighborhood Embedding and Weighting for Sparse Point Clouds Segmentation
• AccSS3D: Accelerator for Spatially Sparse 3D DNNs
• Verifiability and Predictability: Interpreting Utilities of Network Architectures for Point Cloud Processing
• Learn to Learn Metric Space for Few-Shot Segmentation of 3D Shapes
• SK-Net: Deep Learning on Point Cloud via End-to-end Discovery of Spatial Keypoints
• Dense Graph Convolutional Neural Networks on 3D Meshes for 3D Object Segmentation and Classification