This package takes inputs from a 3D lidar and separate camera, combines the inputs in such a way as to produce two separate outputs:
- A PointCloud2, that includes all the LIDAR points that overlap the camera image with added R, G and B data channels that contain the color of the point in the environment.
- A standard uint16 depth image that is of the same size and FOV as the camera image and has metric depth data for each of the pixels.
ROS distribution: Jade Ubuntu: Ubuntu 14.04
Since there were conflicts between the opencv versions(ROS Jade has dependencies to opencv2.4) cv_bridge and image_pipeline source code was cloned into the workspace to be recompiled.
Before running the roslaunch command make sure that you have created a "bag_files" folder in the lidar_camera package and a rosbag "bag1.bag" file is inside.
Run the following command to launch the nodes:
roslaunch lidar_camera combine_lidar_camera.launch
This will start the rosbag, combine_lidar_camera node and image_view for displaying the depth image.
Filtered and colored pointcloud:
Depth image:
