Pointwise Convolutional Neural Networks

Binh-Son Hua, Minh-Khoi Tran, Sai-Kit Yeung
12/14/2017
cs.CVcs.LG

Official code: hkust-vgd/pointwise (132 stars, MIT, last commit 2020-01-02).

Abstract

Deep learning with 3D data such as reconstructed point clouds and CAD models has received great research interests recently. However, the capability of using point clouds with convolutional neural network has been so far not fully explored. In this paper, we present a convolutional neural network for semantic segmentation and object recognition with 3D point clouds. At the core of our network is pointwise convolution, a new convolution operator that can be applied at each point of a point cloud. Our fully convolutional network design, while being surprisingly simple to implement, can yield competitive accuracy in both semantic segmentation and object recognition task.

View on arXivView PDF

Code Implementations(1)

hkust-vgd/pointwiseOfficial100%
MIT

Code for Pointwise Convolutional Neural Networks, CVPR 2018

13225C++, CudaApr 3, 20186 years agoMIT
deep-learningdeep-neural-networkspoint-cloud

Similar Papers

A comprehensive overview of deep learning techniques for 3D point cloud classification and semantic segmentation

85% match2024

LatticeNet: Fast Point Cloud Segmentation Using Permutohedral Lattices

85% match1 repo482019

Dynamic Graph CNN for Learning on Point Clouds

84% match1 repo1.9k2018

PointConv: Deep Convolutional Networks on 3D Point Clouds

84% match1 repo4982018

Human Vision Based 3D Point Cloud Semantic Segmentation of Large-Scale Outdoor Scene

84% match1 repo162023

TransUPR: A Transformer-based Uncertain Point Refiner for LiDAR Point Cloud Semantic Segmentation

84% match2023

Cite this paper

@article{hua2017pointwise,
  title  = {Pointwise Convolutional Neural Networks},
  author = {Binh-Son Hua and Minh-Khoi Tran and Sai-Kit Yeung},
  year   = {2017},
  eprint = {1712.05245},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1712.05245}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1712.05245.svg)](https://codeofpaper.com/paper/1712.05245)

Discussion