Local Spectral Graph Convolution for Point Set Feature Learning

By Chu Wang, Babak Samari, Kaleem Siddiqi. 2018

401 citations recorded by the source corpus.

Feature learning on point clouds has shown great promise, with the introduction of effective and generalizable deep learning frameworks such as pointnet++. Thus far, however, point features have been abstracted in an independent and isolated manner, ignoring the relative layout of neighboring points as well as their features. In the present article, we propose to overcome this limitation by using spectral graph convolution on a local graph, combined with a novel graph pooling strategy. In our approach, graph convolution is carried out on a nearest neighbor graph constructed from a point's neighborhood, such that features are jointly learned. We replace the standard max pooling step with a recursive clustering and pooling strategy, devised to aggregate information from within clusters of nodes that are close to one another in their spectral coordinates, leading to richer overall feature descriptors. Through extensive experiments on diverse datasets, we show a consistent demonstrable advantage for the tasks of both point set classification and segmentation.

Subfields: Computational Mechanics, Environmental Engineering, Geology

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

In Environmental Engineering

In Geology

Citations in the corpus

Revisiting Point Cloud Classification: A New Benchmark Dataset and Classification Model on Real-World Data (2019, methodology): “…[14] - 90.2 Kd-Net [23] 88.5 90.6 (91.8 *) PointNet++ [32] 87.8 90.7 (91.9 w/ normal) SO-Net [24] 87.3 90.9 (93.4 w/ normal) KCNet [33] - 91 3DmFV [2] 86.3 91.4 SpecGCN [41] - 91.5 (92.1 w/ normal) SpiderCNN [44] 86.8 90.0 (92.4 w/ normal) DGCNN [42] 90.2 92.2 PointCNN [25] 88.8 92.5 Table 1.”
PointCNN: Convolution On X-Transformed Points (2018, methodology): “…83.3 - - SPLATNet [43] 85.4 83.7 - - SpiderCNN [53] 85.3 81.7 - - SO-Net [27] 84.9 81.0 - - PCNN [3] 85.1 81.8 - - KCNet [42] 83.7 82.2 - - SpecGCN [46] 85.4 - - - Kd-Net [22] 82.3 77.4 - - 3DmFV-Net [4] 84.3 81.0 - - RSNet [18] 84.9 81.4 56.47 - DGCNN [50] 85 We note that PointCNN outperforms all…”
PointWeb: Enhancing Local Neighborhood Features for Point Cloud Processing (2019, neutral): “9 SpecGCN [30] point - 92.”
PointContrast: Unsupervised Pre-training for 3D Point Cloud Understanding (2020, background): “Second, is the choice of neighborhood aggregation mechanism which could either be hierarchical [48, 33, 82, 16, 35], spatial CNN-like [30, 73, 37, 85, 59], spectral [78, 62, 65] or graph-based [72, 64, 68, 54].”
Geometric Back-Projection Network for Point Cloud Classification (2019, neutral): “ 92.8 RS-CNN [18] coords 1k - 92.9 DGCNN [43] coords 1k 90.2 92.9 KP-Conv [34] coords 1k - 92.9 Ours coords 1k 91.0 93.8 SO-Net [12] coords 2k 87.3 90.9 PointNet++ [27] coords+norm 5k - 91.9 Spec-GCN [39] coords+norm 2k - 92.1 SpiderCNN [47] coords+norm 5k - 92.4 DensePoint [17] coords+voting 1k - 9”
Transductive Zero-Shot Learning for 3D Point Cloud Classification (2019, methodology): “Any network that learns a feature space from 3D point sets and is invariant to permutations of points in the point cloud can be used in our method as the point cloud network [30, 31, 55, 24, 61, 54, 60].”
Revisiting Point Cloud Shape Classification with a Simple and Effective Baseline (2021, methodology): “Hence recent methods, including RSCNN and DGCNN, do not use it.”
Learning Object Bounding Boxes for 3D Instance Segmentation on Point Clouds (2019, background): “To address it, a number of approaches [38; 57; 43; 31; 55; 49; 26; 17] have been proposed recently.”
Mitigating the Hubness Problem for Zero-Shot Learning of 3D Objects (2019, methodology): “In the left branch, a feature vector κ(X ) ∈ Rm from a point cloud X is extracted using a point cloud network, which can be any network that learns a feature space from 3D point cloud data with the capability of being invariant to permutations in the point cloud [19, 26, 27, 37, 38, 43].”
KPConv: Flexible and Deformable Convolution for Point Clouds (2019, neutral): “SpecGCN [40] 91.”
Deep Learning for LiDAR Point Clouds in Autonomous Driving: A Review (2020, background): “Those graph CNNs define convolutions directly on the graph in the spectral and non-spectral (spatial) domain, operating on groups of spatially close neighbors [84].”
PAConv: Position Adaptive Convolution with Dynamic Kernel Assembling on Point Clouds (2021, background): “Follow-up works [37, 38, 48, 20, 51] exploit local aggregation schemes to improve the network.”
Graph convolutional networks: a comprehensive review (2019, background): “propose a local spectral graph convolutional network for both point cloud classification and segmentation [90].”
Group-Free 3D Object Detection via Transformers (2021, background): “Network architecture for Point Cloud A large set of network architectures [38, 12, 29, 9, 47, 23, 44, 39, 45, 28, 30, 33, 43, 40, 17, 1, 41, 49, 10, 22] have been proposed for various point cloud based learning tasks.”
Review: deep learning on 3D point clouds (2020, methodology): “LocalSpec [45] Farthest point sampling k-NN Spectral convolution+cluster pooling”
Spatial Information Guided Convolution for Real-Time RGBD Semantic Segmentation (2020, background): “There are other improvements [39]–[41] on utilizing neural networks to effectively extract deep features from 3D point sets.”