ROS package implementing local pipeline for 3D object recognition from the paper A. Aldoma et al., "Tutorial: Point Cloud Library: Three-Dimensional Object Recognition and 6 DOF Pose Estimation," in IEEE Robotics & Automation Magazine, vol. 19, no. 3, pp. 80-91, Sept. 2012.
Compiling catkin workspase:
cd ~/catkin_ws
catkin_make
source devel/setup.bash
Running recognition node:
rosrun local_object_recognizer recognizer _models_dir:=<models_dir> _gt_dir:=<ground_truth_dir> <options>
ground_truth_dir - Path to ground truth files.
Options are:
- _h=true: Show help
- _scene_pcd:=<cloud.pcd> Scene point cloud (normal mode only)
- _model: Model name to be used instead of reading models file
- _ransac:=true Use RANSAC in ICP step
- _rf_radius:=<rf_radius> RF radius
- _cg_size:=<cg_size> Correspondence grouping size
- _cg_thresh:=<cg_thresh> Correspondence grouping threshold
- _vis:=true Visualize recognition results