PPFNet: Global Context Aware Local Features for Robust 3D Point Matching
H. Deng, T. Birdal, and S. Ilic. (2018)cite arxiv:1802.02669Comment: Accepted for publication at CVPR 2018.
Abstract
We present PPFNet - Point Pair Feature NETwork for deeply learning a globally
informed 3D local feature descriptor to find correspondences in unorganized
point clouds. PPFNet learns local descriptors on pure geometry and is highly
aware of the global context, an important cue in deep learning. Our 3D
representation is computed as a collection of point-pair-features combined with
the points and normals within a local vicinity. Our permutation invariant
network design is inspired by PointNet and sets PPFNet to be ordering-free. As
opposed to voxelization, our method is able to consume raw point clouds to
exploit the full sparsity. PPFNet uses a novel $N-tuple$ loss and
architecture injecting the global information naturally into the local
descriptor. It shows that context awareness also boosts the local feature
representation. Qualitative and quantitative evaluations of our network suggest
increased recall, improved robustness and invariance as well as a vital step in
the 3D descriptor extraction performance.
Description
[1802.02669] PPFNet: Global Context Aware Local Features for Robust 3D Point Matching
%0 Journal Article
%1 deng2018ppfnet
%A Deng, Haowen
%A Birdal, Tolga
%A Ilic, Slobodan
%D 2018
%K 3dmatch descriptor pointnet pointpairfeatures registration shotlrf tupleloss
%T PPFNet: Global Context Aware Local Features for Robust 3D Point Matching
%U http://arxiv.org/abs/1802.02669
%X We present PPFNet - Point Pair Feature NETwork for deeply learning a globally
informed 3D local feature descriptor to find correspondences in unorganized
point clouds. PPFNet learns local descriptors on pure geometry and is highly
aware of the global context, an important cue in deep learning. Our 3D
representation is computed as a collection of point-pair-features combined with
the points and normals within a local vicinity. Our permutation invariant
network design is inspired by PointNet and sets PPFNet to be ordering-free. As
opposed to voxelization, our method is able to consume raw point clouds to
exploit the full sparsity. PPFNet uses a novel $N-tuple$ loss and
architecture injecting the global information naturally into the local
descriptor. It shows that context awareness also boosts the local feature
representation. Qualitative and quantitative evaluations of our network suggest
increased recall, improved robustness and invariance as well as a vital step in
the 3D descriptor extraction performance.
@article{deng2018ppfnet,
abstract = {We present PPFNet - Point Pair Feature NETwork for deeply learning a globally
informed 3D local feature descriptor to find correspondences in unorganized
point clouds. PPFNet learns local descriptors on pure geometry and is highly
aware of the global context, an important cue in deep learning. Our 3D
representation is computed as a collection of point-pair-features combined with
the points and normals within a local vicinity. Our permutation invariant
network design is inspired by PointNet and sets PPFNet to be ordering-free. As
opposed to voxelization, our method is able to consume raw point clouds to
exploit the full sparsity. PPFNet uses a novel $\textit{N-tuple}$ loss and
architecture injecting the global information naturally into the local
descriptor. It shows that context awareness also boosts the local feature
representation. Qualitative and quantitative evaluations of our network suggest
increased recall, improved robustness and invariance as well as a vital step in
the 3D descriptor extraction performance.},
added-at = {2018-12-18T10:26:57.000+0100},
author = {Deng, Haowen and Birdal, Tolga and Ilic, Slobodan},
biburl = {https://www.bibsonomy.org/bibtex/2a92b7b0d0312fd7abb8700636519390b/rspezialetti},
description = {[1802.02669] PPFNet: Global Context Aware Local Features for Robust 3D Point Matching},
interhash = {95a02483bea735b79e424669a2fc3358},
intrahash = {a92b7b0d0312fd7abb8700636519390b},
keywords = {3dmatch descriptor pointnet pointpairfeatures registration shotlrf tupleloss},
note = {cite arxiv:1802.02669Comment: Accepted for publication at CVPR 2018},
timestamp = {2018-12-18T10:26:57.000+0100},
title = {PPFNet: Global Context Aware Local Features for Robust 3D Point Matching},
url = {http://arxiv.org/abs/1802.02669},
year = 2018
}