Package Summary
Segmentation of 3d pointclouds in clutter vs. table surface Uses classifiers trained from 104 hand-labeled clutter table scans combing 3d LIDAR and camera color information. The original training data is at: www.hrl.gatech.edu/data/clutter Brunt of the work is done by: processor.py Main file is: run_segmentation_PR2.py
- Author: Jason Okerman, Martin Schuster, Advisors: Prof. Charlie Kemp and Jim Regh, Lab: Healthcare Robotics Lab at Georgia Tech
- License: BSD
- Repository: gt-ros-pkg
- Source: svn http://gt-ros-pkg.googlecode.com/svn/trunk/hrl/clutter_svm_segmentation
Contents
gt-ros-pkg: clutter_sgm_segmentation.
Data Set
This package segments a point cloud into 'clutter' and 'surface' by combining the 3D information with a camera image. Classifiers were trained from 100+ cluttered tables scans.
The original training dataset is available at hrl.gatech.edu/data/clutter