search

acfr / cam_lidar_calibration

(ITSC 2021) Optimising the selection of samples for robust lidar camera calibration. This package estimates the calibration parameters from camera to lidar frame.
Apache License 2.0
436 stars 102 forks source link

Subtract Background to get Board Dimensions #36

Closed jclinton830 closed 1 year ago

jclinton830 commented 1 year ago

New method allows

[-] Use the feature extraction ui to create a region of interest where the scene is static [-] Take a sample of the static scene [-] The captured sample is subtracted from every consecutive frame to then extract any other new item in the scene. This is used to distinguish just the board in the frame [-] Capture sample - Now takes 5 consecutive frames and gets the running average of the board's parameters [-] Take as many samples as you need [-] Optimise does what it used to do

Other changes

[-] Visualise Results script takes care of angle wrapping to stop sudden angle sign fluctuations at +/-2Pi [-] Pre-commit is set up [-] Enforce Clang to format all the files appropriately, was hard to read and needed consistency very badly.

What's next?

@chinitaberrio needs to test with our car and sensors. Once this is done we can merge it.

juliangaal commented 1 year ago

The diff is very large, to the point of being almost impossible to understand. You included multiple data samples in data/. Was this intentional?

jclinton830 commented 1 year ago

The diff is very large, to the point of being almost impossible to understand. You included multiple data samples in data/. Was this intentional?

Good pick up. I did not notice them being added too. I will get rid of them.

jclinton830 commented 1 year ago

We have decided not to merge this with the current master branch. We will rather be making this our default branch and the master will still be called master but will contain the original code presented in the paper.