New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Stereo Camera - nerian/right/camera_info/P matrix Tx value #12
Comments
Hi, @Prasheel24, Thank you for your interest in RELLIS-3D. Hope this helps! |
Hi, Prasheel,
We don't have a separate bag file, but we are trying to find a way that allows people to use the calibration results with the rosbag. But I don't know when we will be able to make it. I will appreciate that if you can contribute your solution. Thanks!
…________________________________
发件人: Prasheel Renkuntla ***@***.***>
发送时间: 2021年7月14日 19:45
收件人: unmannedlab/RELLIS-3D ***@***.***>
抄送: maskjp ***@***.***>; Mention ***@***.***>
主题: Re: [unmannedlab/RELLIS-3D] Stereo Camera - nerian/right/camera_info/P matrix Tx value (#12)
Thanks @maskjp<https://github.com/maskjp>!
Do you have a separate bag file in an outdoor scenario with these camera parameters published.
I am not quite sure if publishing them on a separate topic would not conflict with the timestamps. As far as I have seen, RTABMAP does not let these parameters set in a launch file but instead subscribes onto a topic to be consistent with timestamps. I am looking for workarounds and will keep you posted. But please do let me know if you come across such rosbags.
―
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub<#12 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ACW7NVBOOEJWT5CIN7LB6XLTXXSLZANCNFSM5AKALTJQ>.
|
Thank you. I republished these parameters from a yaml file onto a new topic (hoping only the D,K,R,P were different). Now I am facing another issue though which I have posted on the rtabmap forum. But, please do let me know if my consideration of the new calibrated params is correct, i.e, only these are republished from each of the cameras- M, D, R, P.
This is my publisher-
Also, I saw that the header frame_id for topic /nerian/right/camera_info was nerian_left. Is it the parent frame? Thanks, |
This is by convention for stereo cameras, as described here. The We have also written a small script to that publishes the new |
Hi, @Prasheel24, I pushed the script that @phil0stine wrote here. Please take a look at it. You can run the following command to test the script and use rviz or stereo_view to visualize the results.
Best wishes! |
Hello Team,
First of all, I would like to appreciate your work on the Data set! Really useful for the tasks that we want to do.
However, I have failed trying to use the Full Stack Merged (00000_20210224.zip) with a ROS pkg called RTABMAP.
My task is to develop an occupancy grid from the outdoor scenario which your data provides. In RTABMAP, I am following this tutorial -Stereo Outdoor Mapping which works well on their dataset. I believe the issue might be that the nerian left and right camera are not calibrated well.
Following the tutorial(Stereo Calibration), the P matrix elements for the left and right cameras should be something like -
left/camera_info/P : [487.608731712861, 0.0, 318.1162109375, 0.0, 0.0, 487.608731712861, 249.44425201416, 0.0, 0.0, 0.0, 1.0, 0.0]
right/camera_info/P : [487.608731712861, 0.0, 318.1162109375, -58.3626989865376, 0.0, 487.608731712861, 249.44425201416, 0.0, 0.0, 0.0, 1.0, 0.0]
Because, RTABMAP assumes a horizontal left/right stereo setup where the Tx (or P(0,3)) is negative in the right camera info msg, i.e the stereo baseline (0.000000) should be positive (baseline=-Tx/fx).
But from the Full Stack merged data, I get the following
/nerian/left/camera_info/P: [635.618713, 0.0, 414.684479, 0.0, 0.0, 635.954285, 292.338189, 0.0, 0.0, 0.0, 1.0, 0.0]
/nerian/right/camera_info/P: [639.706055, 0.0, 388.850254, 0.0, 0.0, 641.632874, 290.169922, 0.0, 0.0, 0.0, 1.0, 0.0]
I have read their forums where people faced a similar issue but as far as I can see, they got it resolved as the right camera Tx value was negative itself and had to perform some minor changes.
I believe this might be the issue but please do enlighten me for my exposure to working with cameras is limited to only RGB-D. I am currently looking at possible workarounds that I can do but if you could please guide me in some way possible that would be great!
Looking forward to hearing from you guys.!
The text was updated successfully, but these errors were encountered: