M2DP-CPP

C++ implementation of the M2DP algorithm. Original repository can be found here.

Introduction:

Multiview 2D projection (M2DP) is a global descriptor of input point cloud.

 Input:
      data               pcl::PointCloud<pcl::PointXYZ>::Ptr      Point cloud. 
Output:
      desM2DP     Eigen::Matrix<double, 192, 1>         M2DP descriptor of the input cloud data
      A                      Eigen::Matrix<double, 64, 128>      Signature matrix

Details of M2DP can be found in the following paper:

Li He, Xiaolong Wang and Hong Zhang, M2DP: A Novel 3D Point Cloud Descriptor and Its Application in Loop Closure Detection, IROS 2016.

Li He, Dept. of Computing Science, University of Alberta [email protected]

Requirement

• Eigen3
• PCL

usage

Follow the below format to extract features from pointcloud:

#include <m2dp.h>

...
pcl::PointCloud<pcl::PointXYZ>::Ptr cloud(new pcl::PointCloud<pcl::PointXYZ>);
M2DP m2dp(cloud);
Eigen::MatrixXd desM2dp;
desM2dp = m2dp.get_m2dp_result();
Eigen::MatrixXd A;
A = m2dp.get_m2dp_A();

Reference

	@inproceedings{he2016m2dp,
  title={M2DP: A novel 3D point cloud descriptor and its application in loop closure detection},
  author={He, Li and Wang, Xiaolong and Zhang, Hong},
  booktitle={2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
  pages={231--237},
  year={2016},
  organization={IEEE}
}

Copyright

See LICENSE for details.