Follow this process to install libfreenect2 and iai_kinect2.
catkin build -DCMAKE_BUILD_TYPE=”Release” -Dfreenect2_DIR=~/src/freenect2/lib/cmake/freenect2
roscore
roslaunch kinect2_bridge kinect2_bridge.launch base_name:=kinect_center
roslaunch kinect_bringup kinect_center_streaming.launch
roslaunch kinect2_bridge kinect2_bridge.launch base_name:=kinect_center depth_method:=cpu
rosrun kinect_bringup tf_interactive_marker.py world kinect_test 1.4784 0.0693 0.4358 0.046 0.4315 -3.1227
1.launch gazebo and rviz
roslaunch kinect_bringup empty_world.launch
1.5. delete ground plane (manual)
2.turn on gravity on gazebo
rosservice call /gazebo/set_physics_properties "
time_step: 0.001
max_update_rate: 1000.0
gravity:
x: 0.0
y: 0.0
z: 0.0
ode_config:
auto_disable_bodies: False
sor_pgs_precon_iters: 0
sor_pgs_iters: 50
sor_pgs_w: 1.3
sor_pgs_rms_error_tol: 0.0
contact_surface_layer: 0.001
contact_max_correcting_vel: 100.0
cfm: 0.0
erp: 0.2
max_contacts: 20"
3.spown the sensor on gazebo and rviz
roscd kinect_bringup/urdf/models
rosrun gazebo_ros spawn_model -urdf -file kinect.urdf -model kinect -x 1.0 -y 0.0 -z 0.5 -R 0.0 -P 0.35 -Y 3.14
rosrun kinect_bringup tf_camera.py world kinect 1 0 0.5 0 0.35 3.14
4.spown target object you want to collect for datasets
roscd kinect_bringup/urdf/models
rosrun gazebo_ros spawn_model -urdf -file object.urdf -model kobject -x 0 -y 0 -z 0.05 -R 0.0 -P 0.582 -Y 0.0
5.generating object pose automatically
rosrun kinect_bringup random_move_euler.py
or
rosrun kinect_bringup tf_state_publisher.py world kobject 0.1 0.1 0.1 0 0 0
6.record object's point cloud and pose
roscd kinect_bringup/data
mkdir data1 ; cd data1
mkdir learn_data
rosrun kinect_bringup record_euler input:=/kinect/sd/points number_of_data
7.normalizing point cloud and make voxel
roscd kinect_bringup/data/data1
rosrun kinect_bringup make_voxel_data number_of_data
8.save dataset in hdf5 format
roscd kinect_bringup/data/data1
rosrun kinect_bringup from_csv2hdf.py -n number of dataset
roslaunch kinect_bringup object.launch
rosrun kinect_bringup tf_interactive_marker_object.py world object_origin 0.1 0.1 0.1 0 0 0
rosrun tf tf_echo estimated_tf object_origin
1.launch gazebo and rviz
roslaunch kinect_bringup empty_world.launch
1.5. delete ground plane (manual)
2.turn on gravity on gazebo
rosservice call /gazebo/set_physics_properties "
time_step: 0.001
max_update_rate: 1000.0
gravity:
x: 0.0
y: 0.0
z: 0.0
ode_config:
auto_disable_bodies: False
sor_pgs_precon_iters: 0
sor_pgs_iters: 50
sor_pgs_w: 1.3
sor_pgs_rms_error_tol: 0.0
contact_surface_layer: 0.001
contact_max_correcting_vel: 100.0
cfm: 0.0
erp: 0.2
max_contacts: 20"
3.delete grand_plane in gazebo
4.spown the sensor on gazebo and rviz
roscd kinect_bringup/urdf/models
rosrun gazebo_ros spawn_model -urdf -file kinect.urdf -model kinect -x 0 -y 0 -z 0.5 -R 0 -P 1.5708 -Y 0
rosrun kinect_bringup tf_camera.py world kinect 0 0 0.5 0 1.5708 0
5.spown target object you want to collect for datasets
roscd kinect_bringup/urdf/models
rosrun gazebo_ros spawn_model -urdf -file hv8.urdf -model kobject -x 0 -y 0 -z 0 -R 0.0 -P 0.0 -Y 0.0
5.save model cloud at kinect_bringup/pcds
rosrun kinect_bringup record_pc