estimateCameraParams with partially occluded patterns

1 view (last 30 days)
I'm trying to use estimateCameraParams to calibrate a single camera. I have several images of my calibration pattern and the corresponding image and object points for each one. However, some of the images have parts of the pattern occluded, where it's not a perfect rectangle, or there might be a hole in some images but not in others. Regardless, every set of corresponding image and object points are consistent with respect to a common origin.
The imagePoints parameter of estimateCameraParams expects an input matrix of shape Mx2xnumImages where M is the number of key point correspondences in each pattern.
In my case, M changes for every image. I don't believe there's an algorithmic limitation since OpenCV's calibrateCamera function allows for M to change.
Is there a way, in MATLAB, to use estimateCameraParams with varying M ?

Answers (0)

Products


Release

R2018a

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!