A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf
.
Filters
Learning Graph-Convolutional Representations for Point Cloud Denoising
[article]
2020
arXiv
pre-print
We propose a deep neural network based on graph-convolutional layers that can elegantly deal with the permutation-invariance problem encountered by learning-based point cloud processing methods. ...
The network is fully-convolutional and can build complex hierarchies of features by dynamically constructing neighborhood graphs from similarity among the high-dimensional feature representations of the ...
This material is based upon work supported by Google Cloud. ...
arXiv:2007.02578v1
fatcat:zv2dyootkvasfg6txymp6fn2z4
Classification of Typical Static Objects in Road Scenes Based on LO-Net
2024
Remote Sensing
Despite the popularity of the PointNet++ network for direct point cloud processing, it encounters issues related to insufficient feature learning and low accuracy. ...
Finally, it employs a point cloud spatial pyramid joint pooling module, developed by the authors, for the multiscale pooling of final low-level local features. ...
Acknowledgments: I would like to thank Haiyang Lv from the School of Geographic and Biologic Information at Nanjing University of Posts and Telecommunications for providing valuable feedback on the paper ...
doi:10.3390/rs16040663
fatcat:47wjoaanmbauzfiy6e4sc5sxli
Octree guided CNN with Spherical Kernels for 3D Point Clouds
[article]
2019
arXiv
pre-print
We propose an octree guided neural network architecture and spherical convolutional kernel for machine learning from arbitrary 3D point clouds. ...
We exploit this association to avert dynamic kernel generation during network training that enables efficient learning with high resolution point clouds. ...
We also thank NVIDIA corporation for donating the Titan XP GPU used in our experiments. ...
arXiv:1903.00343v1
fatcat:lvkqi6izz5e4feslxnzpmxu3fu
Multi-resolution deep learning pipeline for dense large scale point clouds
[article]
2021
arXiv
pre-print
The main challenge of processing such large point clouds remains in the size of the data, which induce expensive computational and memory cost. ...
In this paper, we introduce a new generic deep learning pipeline to exploit the full precision of large scale point clouds, but only for objects that require details. ...
FCPN [RWS * 18] uses both voxel and MLP based networks in a fully-convolutional point network able to process clouds with up to 200k points. ...
arXiv:2109.11311v1
fatcat:bvclol36ojawjk3qtfgviqfl5m
Continuous Conditional Random Field Convolution for Point Cloud Segmentation
2021
Pattern Recognition
Therefore, we first model the point cloud features with a continuous quadratic energy model and formulate its solution process as a message-passing graph convolution, by which it can be easily integrated ...
Point cloud segmentation is the foundation of 3D environmental perception for modern intelligent systems. ...
Natural Science Foundation of China under Grant Nos 61872188, 61703209, U1713208, 61972204, 61672287, 61861136011, 61773215, and by the French Labex MS2T ANR-11-IDEX-0004-02 through the program Investments for ...
doi:10.1016/j.patcog.2021.108357
fatcat:eyz6xdy5dzf3npmcflx4zdhvtu
Beyond single receptive field: A receptive field fusion-and-stratification network for airborne laser scanning point cloud classification
2022
ISPRS journal of photogrammetry and remote sensing (Print)
With a novel dilated graph convolution (DGConv) and its extension annular dilated convolution (ADConv) as basic building blocks, the receptive field fusion process is implemented with the dilated and annular ...
In this article, for the objective of configuring multi-receptive field features, we propose a novel receptive field fusion-and-stratification network (RFFS-Net). ...
For example, Schmohl and Sörgel [38] propose sparse submanifold convolutional networks (SSCNs) to classify voxelized ALS point clouds point by point. ...
doi:10.1016/j.isprsjprs.2022.03.019
fatcat:tnkv5uuqbngxjlqtsro5yr5a64
Point cloud classification by dynamic graph CNN with adaptive feature fusion
2021
IET Computer Vision
The authors propose a new network based on feature fusion to improve the point cloud classification and segmentation tasks. ...
Point cloud data, as the most basic and important form of representation of 3D images, can accurately and intuitively show the real world. ...
the low-level layer feature, middle-level features and high-level features. ...
doi:10.1049/cvi2.12039
fatcat:onn22iai3rayvlcfqjh277po7i
Graph Signal Processing for Geometric Data and Beyond: Theory and Applications
[article]
2021
arXiv
pre-print
community -- enables processing signals that reside on irregular domains and plays a critical role in numerous applications of geometric data from low-level processing to high-level analysis. ...
., 2D depth images, 3D point clouds, and 4D dynamic point clouds, have found a wide range of applications including immersive telepresence, autonomous driving, surveillance, etc. ...
Theoretically
Nodal-domain Methods
Graph Signal Processing Graph Inference
Data Operator Process
Graph Neural Networks Interpret 4D Dynamic Point Cloud time … Fig. 1 : Illustration of GSP for ...
arXiv:2008.01918v3
fatcat:54ankltzznerpo5t5p3lkezvzu
Deep Learning for 3D Point Clouds: A Survey
[article]
2020
arXiv
pre-print
However, deep learning on point clouds is still in its infancy due to the unique challenges faced by the processing of point clouds with deep neural networks. ...
To stimulate future research, this paper presents a comprehensive review of recent progress in deep learning methods for point clouds. ...
Graph-based Methods Graph-based networks consider each point in a point cloud as a vertex of a graph, and generate directed edges for the graph based on the neighbors of each point. ...
arXiv:1912.12033v2
fatcat:qiiyvvuulfccxaiihf2mu23k34
RGCNN: Regularized Graph CNN for Point Cloud Segmentation
[article]
2018
arXiv
pre-print
In this paper, we instead propose a regularized graph convolutional neural network (RGCNN) that directly consumes point clouds. ...
Leveraging on spectral graph theory, we treat features of points in a point cloud as signals on graph, and define the convolution over graph by Chebyshev polynomial approximation. ...
In order to address the above problems, we propose a regularized graph convolutional neural network (RGCNN) for point cloud segmentation. ...
arXiv:1806.02952v1
fatcat:fdxhxw3eivcdrlgg2c74a425ua
IAGC: Interactive Attention Graph Convolution Network for Semantic Segmentation of Point Clouds in Building Indoor Environment
2022
ISPRS International Journal of Geo-Information
Point-based networks have been widely used in the semantic segmentation of point clouds owing to the powerful 3D convolution neural network (CNN) baseline. ...
Most of the current methods resort to intermediate regular representations for reorganizing the structure of point clouds for 3D CNN networks, but they may neglect the inherent contextual information. ...
Acknowledgments: The authors would like to thank the anonymous reviewers and editors for their valuable comments. We also thank the students who participated in the project. ...
doi:10.3390/ijgi11030181
fatcat:kwlfkyxxczappe2hsyd545rk2u
Unsupervised Segmentation for Terracotta Warrior with Seed-Region-Growing CNN(SRG-Net)
[article]
2021
arXiv
pre-print
There are few pieces of researches concentrating on unsupervised point cloud part segmentation. In this paper, we present SRG-Net for 3D point clouds of terracotta warriors to address these problems. ...
Then we present a supervised segmentation and unsupervised reconstruction networks to learn the characteristics of 3D point clouds. ...
As to convolution-based method. RS-CNN takes a local point cloud subset as its input and maps the low-level relation to the high-level relation to learn the feature better. ...
arXiv:2107.13167v1
fatcat:c6xqkf5o2vaodm44cp3umcisoq
AGCN: Adversarial Graph Convolutional Network for 3D Point Cloud Segmentation
2021
British Machine Vision Conference
In this paper, we propose an Adversarial Graph Convolutional Network for 3D point cloud segmentation. ...
3D point cloud segmentation provides a high-level semantic understanding of object structure that is valuable in applications such as medicine, robotics and self-driving. ...
Model Analysis
Conclusion In this paper, we presented a novel neural network approach for point cloud segmentation. ...
dblp:conf/bmvc/KimA21
fatcat:oegq6iglnnchbhoffgnpfk2vva
Spatial Transformer for 3D Point Clouds
[article]
2021
arXiv
pre-print
Deep neural networks are widely used for understanding 3D point clouds. ...
At each point convolution layer, features are computed from local neighborhoods of 3D points and combined for subsequent processing in order to extract semantic information. ...
on k-NN graphs in the edge convolution [22] for point cloud). ...
arXiv:1906.10887v4
fatcat:dbz7qqcwira4plynko73xrdwoe
Differentiable Convolution Search for Point Cloud Processing
[article]
2021
arXiv
pre-print
As a result, PointSeaNet, a deep network that is sufficient to capture geometric shapes at both convolution level and architecture level, can be searched out for point cloud processing. ...
Exploiting convolutional neural networks for point cloud processing is quite challenging, due to the inherent irregular distribution and discrete shape representation of point clouds. ...
Though methods for for point cloud processing. ...
arXiv:2108.12856v1
fatcat:soai7irtlnektdwnk2hmo2xbyi
« Previous
Showing results 1 — 15 out of 17,342 results