PPFNet: Global Context Aware Local Features for Robust 3D Point Matching

2018/02/07 by Haowen Deng, Tolga Birdal, Deng, Haowen +3 · 11 citations
Earth and Planetary Sciences · Engineering · #3D Shape Modeling and Analysis #3D Surveying and Cultural Heritage #Artificial Intelligence (cs.AI) #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Robotics and Sensor-Based Localization

paper · pdf · doi:10.48550/arxiv.1802.02669

openalex publication_date 2018/02/07 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

We present PPFNet - Point Pair Feature NETwork for deeply learning a globally informed 3D local feature descriptor to find correspondences in unorganized point clouds. PPFNet learns local descriptors on pure geometry and is highly aware of the global context, an important cue in deep learning. Our 3D representation is computed as a collection of point-pair-features combined with the points and normals within a local vicinity. Our permutation invariant network design is inspired by PointNet and sets PPFNet to be ordering-free. As opposed to voxelization, our method is able to consume raw point clouds to exploit the full sparsity. PPFNet uses a novel N-tuple loss and architecture injecting the global information naturally into the local descriptor. It shows that context awareness also boosts the local feature representation. Qualitative and quantitative evaluations of our network suggest increased recall, improved robustness and invariance as well as a vital step in the 3D descriptor extraction performance.

Citations

• A Comprehensive Performance Evaluation of 3D Local Feature Descriptors
• Semi-Supervised Classification with Graph Convolutional Networks
• 3DMatch: Learning Local Geometric Descriptors from RGB-D Reconstructions
• 3D ShapeNets: A Deep Representation for Volumetric Shapes
• Deep Learning of Local RGB-D Patches for 3D Object Detection and 6D Pose\n Estimation

Cited by

• MCI-Net: A Robust Multi-Domain Context Integration Network for Point Cloud Registration
• Geometry-to-Image Synthesis-Driven Generative Point Cloud Registration
• Attention-guided reference point shifting for Gaussian-mixture-based partial point set registration
• ViGG: Robust RGB-D Point Cloud Registration using Visual-Geometric Mutual Guidance
• Text2Loc++: Generalizing 3D Point Cloud Localization from Natural Language
• MambaH-Fit: Rethinking Hyper-surface Fitting-based Point Cloud Normal Estimation via State Space Modelling
• DINOReg: Strong Point Cloud Registration with Vision Foundation Model
• Sampling Network Guided Cross-Entropy Method for Unsupervised Point Cloud Registration
• DualReg: Dual-Space Filtering and Reinforcement for Rigid Registration
• Modality-Aware Feature Matching: A Comprehensive Review of Single- and Cross-Modality Techniques
• MS-DGCNN++: Multi-Scale Dynamic Graph Convolution with Scale-Dependent Normalization for Robust LiDAR Tree Species Classification

Related

• PPF-FoldNet: Unsupervised Learning of Rotation Invariant 3D Local Descriptors
• Mining Point Cloud Local Structures by Kernel Correlation and Graph Pooling
• DH3D: Deep Hierarchical 3D Descriptors for Robust Large-Scale 6DoF Relocalization
• PointNet: Deep Learning on Point Sets for 3D Classification and\n Segmentation
• Mini-PointNetPlus: a local feature descriptor in deep learning model for 3d environment perception