3DLineDetection

by xiaohulugo

xiaohulugo / 3DLineDetection

A simple and efficient 3D line detection algorithm for large scale unorganized point cloud

253 Stars 86 Forks Last release: Not found BSD 2-Clause "Simplified" License 12 Commits 0 Releases

Available items

No Items, yet!

The developer of this repository has not created any items for sale yet. Need a bug fixed? Help with integration? A different license? Create a request here:

3DLineDetection

A simple and efficient 3D line detection algorithm for large scale unorganized point cloud. A conference paper based on this code can be found here https://arxiv.org/abs/1901.02532

Prerequisites:

  1. OpenCV > 2.4.x
  2. OpenMP
  3. No other libs

Usage:

  1. build the project with Cmake
  2. run the code
  3. The default parameters are useful for general cases without tunning(at least for these cases in the experiences of the paper). However, you can also adjust the parameters if the result is not very good.

Performance:

On a computer with Intel Core i5-3550p CPU, the computing time for point clouds with 30M, 20M, 10M, 5M, 2M and 1M points is 130s, 80s, 40s, 20s, 8s and 4s, respectively. image

Feel free to correct my code, if you spotted the mistakes. You are also welcomed to Email me: [email protected]

Notice: Basically, this code works well for dense and accurate point clouds, while poor for noisy ones.

Citation:

Please cite the following paper if this you feel this code helpful.

@article{lu2019fast,
title={Fast 3D Line Segment Detection From Unorganized Point Cloud},
author={Xiaohu, Lu and Yahui, Liu and Kai, Li},
journal={arXiv preprint arXiv:1901.02532},
year={2019},
}

We use cookies. If you continue to browse the site, you agree to the use of cookies. For more information on our use of cookies please see our Privacy Policy.