Only released in EOL distros:   fuerte

This is an interface to point_cloud_mapping. It probably needs to be fixed as Advait hasn't used it since January 08, 2010.

Usage

A README file in the package provides these instructions:

numpy_pc_2009Jun02_181409.pkl: pickle of a 3xN numpy matrix of 3D points. This is a real point cloud of a door using a tilting laser range finder.

python ros_point_clouder.py -h will give some information about the usage.

Tests:

./ros_point_clouder -h will display a help message. Run all tests from bin

1. terminal 1 - ./start_downsampler.sh
   • terminal 2 - ./ros_point_clouder --pc ../data/numpy_pc_2009Jun01_182408.pkl
     • downsampled point cloud will be displayed using mayavi2.
2. terminal 1 - ./start_normal_estimator.sh
   • terminal 2 - ./ros_point_clouder --sphere

     A file "normals_cloud_<timestamp>.pkl" will get created. terminal 2 - ./ros_point_clouder -f ../data/normals_cloud_2009Jun02_202114.pkl --plot

     • surface normals and curvature for synthetic sphere will get displayed.
3. Try 2) with "--pc numpy_pc_2009Jun01_182408.pkl" instead of "--sphere" to plot surface normals and curvature of a real point cloud.