How to choose user object in tabletop_object_detector node?

Question

Hi, I want to use Tabletop_objects_detector with a kinect camera manually without using an arm and manipulation stack. but i don't know how can i set the object type which is tabletop_object_detector should be recognize! as i considered the pr2_object_manipulation_stack there is a service in tabletop_object_detector for giving the goal of detection but i couldn't found out how should i use this service!

is there any tutorial for using tabletop_objects_detector?

Thanks.

---

Originally posted by minoo on ROS Answers with karma: 11 on 2012-02-04

Post score: 1

Answer 1

Hi, There is no tutorial for the same. But if you want to do the detection of the tabletop_objects you may look at the tutorials at http://pointclouds.org/documentation/tutorials/ and read through the below topics

a. Euclidean cluster extraction.

b. Cluster recognition and 6D pose estimation.

I used them straight forwardly for my work of detecting the tabletop_objects.

Hope this helps!

Thanks, Karthik

---

Originally posted by karthik with karma: 2831 on 2012-02-07

This answer was ACCEPTED on the original site

Post score: 0