Few prior works study deep learning on point sets. PointNet by Qi et al. is a
pioneer in this direction. However, by design PointNet does not capture local
structures induced by the metric space points live in, limiting its ability to
recognize fine-grained patterns and generalizability to complex scenes. In this
work, we introduce a hierarchical neural network that applies PointNet
recursively on a nested partitioning of the input point set. By exploiting
metric space distances, our network is able to learn local features with
increasing contextual scales. With further observation that point sets are
usually sampled with varying densities, which results in greatly decreased
performance for networks trained on uniform densities, we propose novel set
learning layers to adaptively combine features from multiple scales.
Experiments show that our network called PointNet++ is able to learn deep point
set features efficiently and robustly. In particular, results significantly
better than state-of-the-art have been obtained on challenging benchmarks of 3D
point clouds.
Description
[1706.02413] PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space
%0 Generic
%1 qi2017pointnet
%A Qi, Charles R.
%A Yi, Li
%A Su, Hao
%A Guibas, Leonidas J.
%D 2017
%K 2017 3D arxiv deep-learning paper point-cloud segmentation stanford
%T PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric
Space
%U http://arxiv.org/abs/1706.02413
%X Few prior works study deep learning on point sets. PointNet by Qi et al. is a
pioneer in this direction. However, by design PointNet does not capture local
structures induced by the metric space points live in, limiting its ability to
recognize fine-grained patterns and generalizability to complex scenes. In this
work, we introduce a hierarchical neural network that applies PointNet
recursively on a nested partitioning of the input point set. By exploiting
metric space distances, our network is able to learn local features with
increasing contextual scales. With further observation that point sets are
usually sampled with varying densities, which results in greatly decreased
performance for networks trained on uniform densities, we propose novel set
learning layers to adaptively combine features from multiple scales.
Experiments show that our network called PointNet++ is able to learn deep point
set features efficiently and robustly. In particular, results significantly
better than state-of-the-art have been obtained on challenging benchmarks of 3D
point clouds.
@misc{qi2017pointnet,
abstract = {Few prior works study deep learning on point sets. PointNet by Qi et al. is a
pioneer in this direction. However, by design PointNet does not capture local
structures induced by the metric space points live in, limiting its ability to
recognize fine-grained patterns and generalizability to complex scenes. In this
work, we introduce a hierarchical neural network that applies PointNet
recursively on a nested partitioning of the input point set. By exploiting
metric space distances, our network is able to learn local features with
increasing contextual scales. With further observation that point sets are
usually sampled with varying densities, which results in greatly decreased
performance for networks trained on uniform densities, we propose novel set
learning layers to adaptively combine features from multiple scales.
Experiments show that our network called PointNet++ is able to learn deep point
set features efficiently and robustly. In particular, results significantly
better than state-of-the-art have been obtained on challenging benchmarks of 3D
point clouds.},
added-at = {2018-07-20T09:13:37.000+0200},
author = {Qi, Charles R. and Yi, Li and Su, Hao and Guibas, Leonidas J.},
biburl = {https://www.bibsonomy.org/bibtex/23fc82f869e8dab17973ba52a6e916b7d/analyst},
description = {[1706.02413] PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space},
interhash = {f598427e493e5d39bd2f1a2dd280a938},
intrahash = {3fc82f869e8dab17973ba52a6e916b7d},
keywords = {2017 3D arxiv deep-learning paper point-cloud segmentation stanford},
note = {cite arxiv:1706.02413},
timestamp = {2018-07-20T09:13:37.000+0200},
title = {PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric
Space},
url = {http://arxiv.org/abs/1706.02413},
year = 2017
}