PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space

Charles R. Qi, Li Yi, Hao Su, Leonidas J. Guibas · NeurIPS 2017 · 2017

arXiv Code

Status

To read

Topics

backbones, sampling

Feeds

First paper · height sampling, P1 · ICCV 2027

Used in

E2: PointNet++ with a learned fusion module, Notebook 01: PointNet and PointNet++ baselines on ForestSemantic-MS

What they did

Applies PointNet recursively on nested partitions of the point set, built with farthest point sampling and ball grouping, and adds multi-scale grouping layers that adapt to varying point density.

My note

No note yet.

Code

Repository

charlesq34/pointnet2

Framework

TensorFlow

Licence

See the repository

My status

Not tried

BibTeX

@inproceedings{qi2017-pointnetpp,
  title = {{PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space}},
  author = {Charles R. Qi and Li Yi and Hao Su and Leonidas J. Guibas},
  booktitle = {NeurIPS 2017},
  year = {2017},
  eprint = {1706.02413},
  archivePrefix = {arXiv}
}