Skip to content
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

T265 distortion matrix #3497

Closed
abylikhsanov opened this issue Mar 14, 2019 · 3 comments
Closed

T265 distortion matrix #3497

abylikhsanov opened this issue Mar 14, 2019 · 3 comments
Labels
T260 series Intel® T265 library

Comments

@abylikhsanov
Copy link

abylikhsanov commented Mar 14, 2019


Required Info
Camera Model { T265
Firmware Version (Open RealSense Viewer --> Click info)

Has anybody calibrated T265 camera? I have tried several times and got this result:

camera matrix
284.509100 0.000000 421.896335
0.000000 282.941856 398.100316
0.000000 0.000000 1.000000

distortion
-0.014216 0.060412 -0.054711 0.011151

rectification
0.999988 -0.003180 -0.003693
0.003180 0.999995 0.000038
0.003693 -0.000049 0.999993

projection
284.509100 0.000000 421.896335 3.910271
0.000000 282.941856 398.100316 0.000000
0.000000 0.000000 1.000000 0.000000

The FoV is being reduced though. But the problem is when I try to run the OpenCV image rectification with those values, the process dies.

Also, I have noticed that the t265 launch file has an argument called "json_file", what is the purpose of it? I am guessing the JSON file for the calibration? If yes, can we have some example of it?

As T265 has 2 cameras, calibration parameters will differ in the projection matrix...

@dorodnic dorodnic added the T260 series Intel® T265 library label Mar 17, 2019
@bfulkers-i
Copy link
Contributor

@abylikhsanov camera intrinsics for T265 are not yet exposed in librealsense, but it is being worked on. If you follow the progress of #3541, you will know as soon as it is ready.

Note that since T265 has a very wide field of view, undistorting the image in OpenCV may result in a very large output image which may crash your machine.

The json file you mention in the ROS wrapper is in fact extrinsic calibration for adding wheel odometry. You can find more information about it in this PR #3462 but this will eventually be replaced with something easier to use.

@bfulkers-i
Copy link
Contributor

Hi @abylikhsanov, intrinsics and extrinsics are now exposed in librealsense 2.20.0, you can see an example using both to render a virtual object in the fisheye frame here:

https://github.com/IntelRealSense/librealsense/tree/master/examples/ar-basic

Can I close this issue?

@abylikhsanov
Copy link
Author

abylikhsanov commented Apr 10, 2019 via email

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
T260 series Intel® T265 library
Projects
None yet
Development

No branches or pull requests

3 participants