Published on Thu Jan 03 2019

GeoNet: Deep Geodesic Networks for Point Cloud Analysis

Tong He, Haibin Huang, Li Yi, Yuqian Zhou, Chihao Wu, Jue Wang, Stefano Soatto

GeoNet is the first deep learning architecture trained to model the intrinsic structure of surfaces represented as point clouds. The method improves the state-of-the-art on multiple representative tasks.

0
0
0
Abstract

Surface-based geodesic topology provides strong cues for object semantic analysis and geometric modeling. However, such connectivity information is lost in point clouds. Thus we introduce GeoNet, the first deep learning architecture trained to model the intrinsic structure of surfaces represented as point clouds. To demonstrate the applicability of learned geodesic-aware representations, we propose fusion schemes which use GeoNet in conjunction with other baseline or backbone networks, such as PU-Net and PointNet++, for down-stream point cloud analysis. Our method improves the state-of-the-art on multiple representative tasks that can benefit from understandings of the underlying surface topology, including point upsampling, normal estimation, mesh reconstruction and non-rigid shape classification.

Wed Jan 24 2018
Computer Vision
Dynamic Graph CNN for Learning on Point Clouds
Point clouds provide a flexible geometric representation suitable for computer graphics. The recent success of convolutional neural networks (CNNs) for image analysis suggests the value of adapting insight from CNN to the point cloud world. We propose a new neural network module dubbed EdgeConv suitable for CNN-based
1
0
0
Sat Dec 05 2020
Computer Vision
ParaNet: Deep Regular Representation for 3D Point Clouds
0
0
0
Wed Aug 04 2021
Computer Vision
Point Discriminative Learning for Unsupervised Representation Learning on 3D Point Clouds
Deep learning has achieved significant progress on point cloud analysis tasks. We propose a point discriminative learning method for unsupervised representation learning on 3D point clouds.
0
0
0
Tue Dec 19 2017
Computer Vision
FoldingNet: Point Cloud Auto-encoder via Deep Grid Deformation
A novel end-to-end deep auto-encoder is proposed to address unsupervised learning challenges on point clouds. A novel folding-based decoder deforms a canonical 2D grid onto the underlying 3D object surface of a point cloud.
0
0
0
Mon Nov 19 2018
Computer Vision
Modeling Local Geometric Structure of 3D Point Clouds using Geo-CNN
Recent advances in deep convolutional neural networks have motivated researchers to adapt CNNs to directly model points in 3D point clouds. We propose Geo-CNN, which applies a generic convolution-like operation dubbed as GeoConv to each point and its local neighborhoods.
0
0
0
Thu Mar 12 2020
Machine Learning
Learning to Segment 3D Point Clouds in 2D Image Space
In this paper we study the problem of how to effectively and efficiently project point clouds into a 2D image space so that traditional 2D convolutional neural networks (CNNs) such as U-Net can be applied for segmentation. With the help of the Delaunay
0
0
0