[ros-users] stereo camera calibration

Asghari Oskoei, Mohammadreza m.asghari-oskoei at herts.ac.uk
Tue Feb 8 08:08:19 UTC 2011

Hi, Thomas, Patrick, joq, all

Thank you for catching, as far as I checked, diamondback binary files still are not supporting -approximate=0.01, and svn trunk does not contain all my required nodes, such as prosilica_node and cob_app. I wonder if I kept working in cturtle and built a node (with cpp lib) which modifies topics' time stamp this could solve the sync problem in stereo images. This new node will subscribe to topics of two cameras (image_raw and camera_info) and publish new topics with same content but the latest time stamp.


From: ros-users-bounces at code.ros.org [mailto:ros-users-bounces at code.ros.org] On Behalf Of Patrick Mihelich
Sent: 04 February 2011 23:15
To: User discussions
Subject: Re: [ros-users] stereo camera calibration

To be specific, once you have image_pipeline trunk overlaid on unstable / diamondback, you can run:

rosrun camera_calibration cameracalibrator.py [other otions] --approximate=0.01

This allows a "slop" of 0.01s between image pairs. If you still don't get a window pop up, you may need to increase the slop. Note: this is because Python message_filters doesn't have an equivalent of the C++ ApproximateTime policy yet, so we're using a simpler algorithm. The other stereo image_pipeline nodes don't require you to set a slop.

Last night I was able to calibrate the narrow stereo pair on a PR2 with their external triggering (for exactly matching timestamps) disabled. But I would greatly appreciate other users trying it out with their hardware.

With the other stereo nodes in image_pipeline, you can enable approximate matching by setting the approximate_sync parameter:

ROS_NAMESPACE=my_stereo rosrun stereo_image_proc stereo_image_proc _approximate_sync:=True

On Fri, Feb 4, 2011 at 3:01 PM, Jack O'Quin <jack.oquin at gmail.com<mailto:jack.oquin at gmail.com>> wrote:
On Fri, Feb 4, 2011 at 2:13 PM, Jack O'Quin <jack.oquin at gmail.com<mailto:jack.oquin at gmail.com>> wrote:
> On Fri, Feb 4, 2011 at 5:23 AM, Thomas Moulard <thomas.moulard at gmail.com<mailto:thomas.moulard at gmail.com>> wrote:
>> On Fri, Feb 4, 2011 at 12:17 PM, Asghari Oskoei, Mohammadreza
>> <m.asghari-oskoei at herts.ac.uk<mailto:m.asghari-oskoei at herts.ac.uk>> wrote:
>>> Hi
>>> I'm using a pair of Procilica cameras on care-o-bot (cob) robot, managed to
>>> calibrate them individually (monocular) but have a problem in stereo
>>> calibration. By running script mentioned for stereo calibration in your
>>> tutorial, I got ok messages for receiving services but no window that might
>>> be shown to calibrate, save or commit.
>>> By the way, by running stereo_image_proc, there are topics in the topic list
>>> (rostopic list), but no output comes out as you check them by rostopic hz or
>>> bw or echo. I have seen this in published topics by stereo_image_proc and
>>> face_detector, though they are in stream or continuous modes.
>> Your camera are probably not synchronized, you can take a look at this
>> topic for more information:
>> https://code.ros.org/lurker/message/20101130.125637.9414e705.en.html
> There is an open enhancement ticket for this:
>  https://code.ros.org/trac/ros-pkg/ticket/4217
> As far as I can tell this is still intended for release in
> Diamondback. But, it is still open and Patrick seems busy with many
> other urgent tickets right now. I don't know what the prognosis is.
Patrick recently committed a fix for #4217!

To test it, you'll need to check out the latest image_pipeline trunk
on top of unstable (or diamondback).


ros-users mailing list
ros-users at code.ros.org<mailto:ros-users at code.ros.org>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ros.org/pipermail/ros-users/attachments/20110208/67411a9a/attachment-0003.html>

More information about the ros-users mailing list