where is the "camera_marker" defined?

[superbigzz]

I dont know where the "camera_marker" is defined. Do I need to define it by myself?

[YanhuiDua]

发自我的华为手机-------- 原始邮件 --------发件人： 冻鱼 <notifications@github.com>日期： 2019年9月8日周日 晚上9:11收件人： IFL-CAMP/easy_handeye <easy_handeye@noreply.github.com>抄送： Subscribed <subscribed@noreply.github.com>主 题： [IFL-CAMP/easy_handeye] where is the "camera_marker" defined? (#47)I dont know where the "camera_marker" is defined. Do I need to define it by myself? —You are receiving this because you are subscribed to this thread.Reply to this email directly, view it on GitHub, or mute the thread.

[marcoesposito1988]

Hi @superbigzz,

That is just a place holder for the actual tf frame published by the ROS driver of your tracking system. Since it is always different, all the launch files of easy_handeye allow you to specify the tf frames of interest via arguments. You can take a look at the readme to see how to include the launch files in your own to calibrate and then use the calibration.

[superbigzz]

Hi @superbigzz,

That is just a place holder for the actual tf frame published by the ROS driver of your tracking system. Since it is always different, all the launch files of easy_handeye allow you to specify the tf frames of interest via arguments. You can take a look at the readme to see how to include the launch files in your own to calibrate and then use the calibration.

oh, got it. Thanks a lot!

[mbajlo]

Hello I am confused with this part:

<arg name="tracking_base_frame" value="/optical_origin"/>
<arg name="tracking_marker_frame" value="/optical_target"/>

Where are tracking base frame and tracking marker frame are coming from?
Since the chessboard or the aruco are on the table, we should somehow estimate their transforms/poses?

[marcoesposito1988]

Hello @mbajlo,

those are the names of the tf frames that your tracking software respectively uses for your camera and marker. These are either configured by you in the launch parameters of the ROS package that you are using to do the tracking, or documented by its developers if they are not configurable.

You don't need to estimate the initial pose. You just have to make sure that the tracking system and its software are properly calibrated and configured (intrinsic camera calibration, stereo camera calibration, marker size and ID, etc.)