Ask Your Question
1

Pose of the object in Roboearth

asked 2011-10-08 09:29:32 -0500

evolutionsdisaster gravatar image

I want to determine the pose of the object that has been detected by Kinect using Roboearth.

I am not able to see the detected node giving me the pose of the object. I did run echo tf, but its giving me someinfo about Kinect and nothing abt the pose of the object. How do we determine this ?? Can any one help me out.

Regards, Faizan Mohammed

edit retag flag offensive close merge delete

5 Answers

Sort by ยป oldest newest most voted
1

answered 2011-11-08 12:50:22 -0500

evolutionsdisaster gravatar image

when you try to save the object's name you would give a class to the object and name to the object. Lets say when you were saying the object you gave the class as "box" and name as "red".

So your object's id is box.red and to get a pose of the object with respect to the kinect you should give the parent id and child id by using rosrun tf tf_echo parent_id child_id.

so here it will be: rosrun tf tf_echo openni_depth_camera box.red

and you will get the pose in terms of translation, quaternions and RPY.

Regards, Faizan Mohammed

edit flag offensive delete link more

Comments

Thanks for the answer!
tfoote gravatar image tfoote  ( 2011-11-08 17:28:09 -0500 )edit
0

answered 2011-11-27 13:10:47 -0500

Sudarshan gravatar image

With the recent update, we have to run rosrun tf tf_echo camera_rgb_optical_frame classname.objectname

I see that camera_rgb_optical_frame and camera_rgb_depth_frame as different nodes of the tree, does that mean they point to different things?

edit flag offensive delete link more
0

answered 2011-11-08 09:31:18 -0500

Sudarshan gravatar image

Could you please post the method you used to get the pose of the object? Thanks.

edit flag offensive delete link more
0

answered 2011-10-08 10:48:45 -0500

evolutionsdisaster gravatar image

I did get the answer on my own, thank you. If any body needs help I can help them out.

edit flag offensive delete link more

Comments

Please post the answer here so that others who find your question will also find your answer.
tfoote gravatar image tfoote  ( 2011-10-08 14:05:45 -0500 )edit
0

answered 2011-10-08 10:20:58 -0500

evolutionsdisaster gravatar image

Well I do see this " re_kinect/detection_results " which would announce the detected object along with the names.

Where is it showing the names ?? When I echo it it gives me jst the point cloud data numbers. And how would the tf work for multiple objects. ??

edit flag offensive delete link more

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer

Question Tools

Stats

Asked: 2011-10-08 09:29:32 -0500

Seen: 583 times

Last updated: Nov 27 '11