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

Issue with rectified image input #40

Open
aditdoshi333 opened this issue Mar 8, 2022 · 5 comments
Open

Issue with rectified image input #40

aditdoshi333 opened this issue Mar 8, 2022 · 5 comments

Comments

@aditdoshi333
Copy link

Hello,

I tried with non rectified images and rectified images. Calibration output is much better with non rectified images. Is the code is expecting distorted images? Are you running rectification in the pipeline anywhere?

Regards

@ChongjianYUAN
Copy link
Member

Hello,

I tried with non rectified images and rectified images. Calibration output is much better with non rectified images. Is the code is expecting distorted images? Are you running rectification in the pipeline anywhere?

Regards

Yes, the input image should be distorted, and the point cloud projection needs to be provided with the same distortion model

@aditdoshi333
Copy link
Author

Okay!

But does the final calibration matrix need to use with the rectified image or non rectified image?

@ChongjianYUAN
Copy link
Member

Okay!

But does the final calibration matrix need to use with the rectified image or non rectified image?

It depends on whether you correlate the LiDAR and camera data through projection, and if so, you need to use the rectified image.

@aditdoshi333
Copy link
Author

I have one more doubt. I am able to calibrate pretty well with your code but not able to use the same matrix in r3live. From the following can you say how is calibration?
157250097-ab28a675-03d6-43bf-813c-7cb31d21d598

@ChongjianYUAN
Copy link
Member

I have one more doubt. I am able to calibrate pretty well with your code but not able to use the same matrix in r3live. From the following can you say how is calibration? 157250097-ab28a675-03d6-43bf-813c-7cb31d21d598

Sorry for the late reply. The extrinsic matrix in livox_camera_calib is from LiDAR coordinate to Camera coordinate and in R3LIVE is from Camera coordinate to LiDAR coordinate. Therefore, you should use the inverse of the output result of livox_camera_calib.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants