ROS Resources: Documentation | Support | Discussion Forum | Index | Service Status | ros @ Robotics Stack Exchange
Ask Your Question

Converting from RGB to depth

asked 2011-02-28 01:50:05 -0500

caspersky gravatar image

updated 2016-10-24 09:00:40 -0500

ngrennan gravatar image


i know that given a (u,v,depth) in the depth image it is easy to retreive the corresponding rgb pixel by unporjecting the depth pixel and projecting to RGB image using the [R T] now my question is, is it possible to do the inverse ? ( having an rgb pixel, retreive the corresponding depth pixel?) and if yes, how ? thx

edit retag flag offensive close merge delete

4 Answers

Sort by ยป oldest newest most voted

answered 2011-03-01 03:16:03 -0500

sebsch gravatar image

openni_camera already provides what you are looking for by default in its current version!

If you have a look at the rgb and depth image you will find that the depth image is already transformed and projected to camera coordinates of the rgb camera: image description

Note that you can turn this off by changing the setting for depth_registration after running:

rosrun dynamic_reconfigure reconfigure_gui

Make sure to have a look at the wiki page as well.

edit flag offensive delete link more

answered 2011-02-28 06:48:58 -0500

caspersky gravatar image

sorry i mentioned this in the tag and not in th post, i am talking about kinect where depth and rgb images are available hope this clarify my question

edit flag offensive delete link more

answered 2011-02-28 04:48:33 -0500

Yogi gravatar image

Sorry, but are you asking if given ONE single image, can we extract depth information? answer is NO.

edit flag offensive delete link more

answered 2011-02-28 21:44:06 -0500

caspersky gravatar image

no answer ?

edit flag offensive delete link more


your questions still does not make complete sense.
Yogi gravatar image Yogi  ( 2011-03-01 01:13:41 -0500 )edit

Question Tools


Asked: 2011-02-28 01:50:05 -0500

Seen: 2,841 times

Last updated: Mar 02 '11