How to see output of tabletop_segmentation

Hi,

I am new to ROS, been going through all tutorials for a week now, the goal is to run plane detection part of tabletopobjectdetector and see how accurate it is wrt detecting the dominant plane. Quite desperate now, almost a week down - little to show for it, so any help - highly appreciated.

I cant seem to make nodes 'click' together.

My setup: XBOX 360 Kinect, 12.04 Precise, Fuerte

Console 1 - roscore
Console 2 - roslaunch openni_launch openni.launch
Console 3 - roslaunch fastplanedetection fastplanedetection_jean.launch
Console 4 - rosrun rviz rviz

All according to documentation, I think.

plane detection node keeps on logging:

[ INFO] [1369276606.899381123]: Waiting for message on topic /narrowstereotextured/left/camera_info
[ INFO] [1369276607.311245337]: Waiting for message on topic /narrowstereotextured/disparity ....

There are no such topics published, however, after launching openni - rostopic list shows following topics:

/camera/depth/camera_info
/camera/depth/disparity
/camera/depth/image
/camera/depth/image/compressed
/camera/depth/image/compressed/parameter_descriptions
/camera/depth/image/compressed/parameter_updates
/camera/depth/image/compressedDepth
... and a bunch of others

So, should I just map /camera/depth/camera_info to /narrowstereotextured/left/camera_info and /camera/depth/camera_info to /narrowstereotextured /disparity ?

And how do I see the output of fastplanedetection in rviz?

Thanks a lot, Greg

Asked by Greg S on 2013-05-22 17:12:03 UTC

Comments

Just a small remark: IIRC, fast_plane_detection has nothing to do with tabletop_object_detector, so you may be looking in the wrong place. Also, the Kinect outputs RGB and depth, the stereo cam outputs disparity. Not sure if conversion from depth to disparity is trivial, never worked with stereo.

Asked by Martin Günther on 2013-05-23 02:19:43 UTC

Thanks, I'll re-post the question with tabletop_detector-only question.

Asked by Greg S on 2013-05-23 11:11:01 UTC

How to see output of tabletop_segmentation

Comments

Answers