depth registered camera_info for ORK

edit

Hi,

I am currently trying to use the Object Recognition Kitchen for detecting an object. However, I am using stereo cameras + stereo_image_proc, and not the standard kinect.

In the tabletop detection pipline tutorial, the configuration file requires

1. A registered (with RGB) depth image topic and
2. depth registered camera info.

For 1) I have seen this. but does someone know how to do 2) , ie, obtain the camera_info for depth registered image ? I have also seen the depth_image_proc package but it doesn't seem to provide anything for this. Is that simply the camera_info for the particular camera (left camera in case of stereo cameras ?) that represents the camera coordinate frame ?

Thanks!