Dynamic Graph CNN for Learning on Point Clouds
Explore this paper's citation graph
Summary
This work proposes a new neural network module suitable for CNN-based high-level tasks on point clouds, including classification and segmentation called EdgeConv, which acts on graphs dynamically computed in each layer of the network.
- Type
- preprint
- Published
- 2018-01-24
- Cited by
- 7,853
- References
- 95
- Access
- Open access
- OpenAlex
- https://openalex.org/W2785053089
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:94822
Keywords
Point cloud, Computer science, Representation (politics), Convolutional neural network, Embedding
References
- Deep Convolutional Networks on Graph-Structured Data
- Laplace-Beltrami eigenfunctions for deformation invariant shape representation
- Multi-view Convolutional Neural Networks for 3D Shape Recognition
- Spectral Networks and Locally Connected Networks on Graphs
- Learning class‐specific descriptors for deformable shapes using localized spectral convolutional networks
- Auto-Encoding Variational Bayes
- Recognizing Objects in 3D Point Clouds with Multi-Scale Local Features
- 3D Object Recognition in Cluttered Scenes with Local Surface Features: A Survey
- Scale-invariant heat kernel signatures for non-rigid shape recognition
- Functional maps
- 3D-Div: A novel local surface descriptor for feature matching and pairwise range image registration
- Towards 3D Point cloud based object maps for household environments
- The wave kernel signature: A quantum mechanical approach to shape analysis
- Aligning point cloud views using persistent feature histograms
- Using Spin Images for Efficient Object Recognition in Cluttered 3D Scenes
- The emerging field of signal processing on graphs: Extending high-dimensional data analysis to networks and other irregular domains
- Shape-based recognition of 3D point clouds in urban environments
- Shape Context: A New Descriptor for Shape Matching and Object Recognition
- The Graph Neural Network Model
- A combined texture-shape descriptor for enhanced 3D feature matching
Cited by
- Adaptive deep learning-based neighborhood search method for point cloud
- Point convolutional neural networks by extension operators
- A survey on deep learning techniques for image and video semantic segmentation
- Dual-Primal Graph Convolutional Networks
- Relational inductive biases, deep learning, and graph networks
- FROM 2D TO 3D SUPERVISED SEGMENTATION AND CLASSIFICATION FOR CULTURAL HERITAGE APPLICATIONS
- PeerNets: Exploiting Peer Wisdom Against Adversarial Attacks
- Monte Carlo convolution for learning on non-uniformly sampled point clouds
- Learning Scene Flow in 3D Point Clouds
- X-Vision: An Augmented Vision Tool with Real-Time Sensing Ability in Tagged Environments
- PointSIFT: A SIFT-like Network Module for 3D Point Cloud Semantic Segmentation
- High Fidelity Semantic Shape Completion for Point Clouds Using Latent Optimization
- Bike flow prediction with multi-graph convolutional networks
- PCN: Point Completion Network
- PVNet: A Joint Convolutional Network of Point Cloud and Multi-View for 3D Shape Recognition
- Object-sensitive potential fields for mobile robot navigation and mapping in indoor environments
- SHPR-Net: Deep Semantic Hand Pose Regression From Point Clouds
- Generating 3D Adversarial Point Clouds
- MeshCNN: a network with an edge
- Graph Neural Networks for IceCube Signal Classification
Related papers
- Continuous Differentiability in the Context of Generalized Approach to Differentiability
- On binary function continuity,differentiability and may differentiability
- The Complexity Of Nowhere Differentiable Continuous Functions
- On derivatives of fuzzy multi-dimensional mappings and applications under generalized differentiability
- On differentiability of Sobolev functions with respect to the Sobolev norm
- There exist no gaps between Gevrey differentiable and nowhere Gevrey differentiable
- HL-differentiability is equivalent to MB#-differentiability
- The approximate solution of variational problems with non-differentiable functionals☆