Dynamic Graph CNN for Learning on Point Clouds
By Yue Wang, Yongbin Sun, Ziwei Liu, Sanjay E. Sarma, Michael M. Bronstein, Justin M. Solomon. 2018
7,693 citations recorded by the source corpus.
Point clouds provide a flexible geometric representation suitable for countless applications in computer graphics; they also comprise the raw output of most 3D data acquisition devices. While hand-designed features on point clouds have long been proposed in graphics and vision, however, the recent overwhelming success of convolutional neural networks (CNNs) for image analysis suggests the value of adapting insight from CNN to the point cloud world. Point clouds inherently lack topological information, so designing a model to recover topology can enrich the representation power of point clouds. To this end, we propose a new neural network module dubbed EdgeConv suitable for CNN-based high-level tasks on point clouds, including classification and segmentation. EdgeConv acts on graphs dynamically computed in each layer of the network. It is differentiable and can be plugged into existing architectures. Compared to existing modules operating in extrinsic space or treating each point independently, EdgeConv has several appealing properties: It incorporates local neighborhood information; it can be stacked applied to learn global shape properties; and in multi-layer systems affinity in feature space captures semantic characteristics over potentially long distances in the original embedding. We show the performance of our model on standard benchmarks, including ModelNet40, ShapeNetPart, and S3DIS.
Subfields: Computational Mechanics, Computer Vision and Pattern Recognition, Graph Learning
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Computational Mechanics
- Leonidas J. Guibas
- Andreas Geiger
- Matthias Nießner
- Daniel Cohen-Or
- Charles Ruizhongtai Qi
- Or Litany
- Yaron Lipman
- Michael Niemeyer
In Computer Vision and Pattern Recognition
- Luc Van Gool
- Dacheng Tao
- Xiaogang Wang
- Ming-Hsuan Yang
- Trevor Darrell
- Jitendra Malik
- Yi Yang
- Andrea Vedaldi
In Graph Learning
- Matthias Fey
- William L. Hamilton
- Michael M. Bronstein
- Stephan Günnemann
- Jure Leskovec
- Jan Eric Lenssen
- Bernard Ghanem
- Peter W. Battaglia
Citations in the corpus
- Weight Standardization (2019, methodology): “We follow the state-of-the-art method DGCNN [41] and use authors’ implementation 2 for all our experiments.”
- Fast Graph Representation Learning with PyTorch Geometric (2019, methodology): “…GCN operator from Schlichtkrull et al. (2018), PointNet++ (Qi et al., 2017), PointCNN (Li et al., 2018), and the continuous kernel-based methods MPNN (Gilmer et al., 2017), MoNet (Monti et al., 2017), SplineCNN (Fey et al., 2018) and the edge convolution operator (EdgeCNN) from Wang et al. (2018b).”
- Relational inductive biases, deep learning, and graph networks (2018, methodology): “…et al., 2015; Gilmer et al., 2017), to predict traffic on roads (Li et al., 2017; Cui et al., 2018), to classify and segment images and videos (Wang et al., 2018c; Hu et al., 2017) and 3D meshes and point clouds (Wang et al., 2018d), to classify regions in images (Chen et al., 2018a), to…”
- SE(3)-Transformers: 3D Roto-Translation Equivariant Attention Networks (2020, neutral): “io/benchmark/): 3DmFV [3], PointNet [19], SpiderCNN [43], PointNet++ [20], DGCN [35].”
- Revisiting Point Cloud Classification: A New Benchmark Dataset and Classification Model on Real-World Data (2019, methodology): “Furthermore, we also introduce BGA-DGCNN , which is a background-aware network based on DGCNN [42].”
- DeepSDF: Learning Continuous Signed Distance Functions for Shape Representation (2019, background): “There is a sizable list of related works for learning on point clouds [37, 51, 56].”
- Learning SO(3) Equivariant Representations with Spherical CNNs (2017, background): “There are two threads within this space, spectral (Bruna et al. 2013b; Defferrard et al. 2016; Kipf and Welling 2016; Yi et al. 2016) and spatial (Boscaini et al. 2016; Masci et al. 2015; Monti et al. 2016; Wang et al. 2018).”
- A Comprehensive Survey on Graph Neural Networks (2019, methodology): “[141], [142], [143] convert point clouds into k-nearest neighbor graphs or superpoint graphs and use ConvGNNs to explore the topological structure.”
- Diffusion Probabilistic Models for 3D Point Cloud Generation (2021, methodology): “Significant progress has been made in developing methods for point cloud analysis, such as classification and segmentation [16, 17, 23].”
- Spherical CNNs on Unstructured Grids (2019, background): “…Segmentation Image semantic segmentation is a classic problem in computer vision, and there has been an impressive body of literature studying semantic segmentation of planar images (Ronneberger et al., 2015; Badrinarayanan et al., 2015; Long et al., 2015; J´egou et al., 2017; Wang et al., 2018a).”
- PointCNN: Convolution On X-Transformed Points (2018, methodology): “[50] proposed to leverage neighborhood structures in both point and feature spaces.”
- Distilling Knowledge From Graph Convolutional Networks (2020, methodology): “Another dataset is ModelNet40 [40] that contains point clouds come from the CAD models, which is a common used dataset for 3D object analysis [28, 30, 19, 37].”
- Point-GNN: Graph Neural Network for 3D Object Detection in a Point Cloud (2020, methodology): “[2] [17] look into classifying a point cloud using a GNN.”
- Campus3D: A Photogrammetry Point Cloud Benchmark for Hierarchical Understanding of Outdoor Scene (2020, methodology): “Three established models are applied: PointNet++ [24], PointCNN [20] and DGCNN [37].”
- DiffusionNet: Discretization Agnostic Learning on Surfaces (2020, methodology): “…[Qi et al. 2017a] and PointNet++ [Qi et al. 2017b] architectures, which have been extended in many recent works, including PointCNN [Li et al. 2018], DGCNN [Wang et al. 2019], PCNN [Atz-mon et al. 2018] and KPConv [Thomas et al. 2019] to name a few (see also [Guo et al. 2020] for a recent survey).”
- 6-DOF GraspNet: Variational Grasp Generation for Object Manipulation (2019, background): “The success of PointNet lead to the introduction of different variations of network architectures [35, 30] that represent 3D data, showing significant improvement on 3D object pose estimation, semantic segmentation, and part segmentation [30, 24, 22, 33].”
- 3D Shape Generation and Completion through Point-Voxel Diffusion (2021, background): “Researchers have explored point cloud classification and segmentation [30, 30, 41] and most assume point cloud processing networks are permutationinvariant.”
- Geometric Back-Projection Network for Point Cloud Classification (2019, methodology): “for a high-level feature representation of every single point, and successfully solved the unordered problem of point cloud with a symmetric function. Due to its effectiveness, later works [23], [46]–[48] also adopted MLP based operations for point cloud processing. Meanwhile, researchers realized t”