Skip to content
A faster implementation of PointNet++ based on PyTorch.
Branch: master
Clone or download
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
pointnet2 code release of Pointnet2.PyTorch Apr 15, 2019
tools bugfixed: use val for validation, support eval mode Apr 16, 2019
.gitignore code release of Pointnet2.PyTorch Apr 15, 2019
LICENSE code release of Pointnet2.PyTorch Apr 15, 2019
README.md Update README.md Apr 15, 2019

README.md

Pointnet2.PyTorch

Installation

Requirements

  • Linux (tested on Ubuntu 14.04/16.04)
  • Python 3.6+
  • PyTorch 1.0

Install

Install this library by running the following command:

cd pointnet2
python setup.py install
cd ../

Examples

Here I provide a simple example to use this library in the task of KITTI ourdoor foreground point cloud segmentation, and you could refer to the paper PointRCNN for the details of task description and foreground label generation.

  1. Download the training data from KITTI 3D object detection website and organize the downloaded files as follows:
Pointnet2.PyTorch
├── pointnet2
├── tools
│   ├──data
│   │  ├── KITTI
│   │  │   ├── ImageSets
│   │  │   ├── object
│   │  │   │   ├──training
│   │  │   │      ├──calib & velodyne & label_2 & image_2
│   │  train_and_eval.py
  1. Run the following command to train and evaluate:
cd tools
python train_and_eval.py --batch_size 8 --epochs 100 --ckpt_save_interval 2 

Project using this repo:

  • PointRCNN: 3D object detector from raw point cloud.

Acknowledgement

You can’t perform that action at this time.