Revision history [back]

I am looking at using the stereoCalibrate function, but there is only one imageSize parameter. I have two receivers attached to a fixed rod and want to calibrate both the intrinsics and extrinsics matrix.

The receivers produce different image sizes, though. E.g. Cam1 is w_1 x h_1 and Cam2 is w_2 x h_2, where w_1 != w_2 and h_1 != h_2.

Given the description of the parameter, though:

imageSize – Size of the image used only to initialize intrinsic camera matrix.

This leads me to believe that it is assumed both images have the same dimensions. Given the goal of computation, I don't believe super-sampling one of the images to make them both the same size is a good decision.

So I should just be calibrating each camera individually?

Thanks for any suggestions!

I am looking at using the stereoCalibrate function, but there is only one imageSize parameter. I have two receivers attached to a fixed rod and want to calibrate both the intrinsics and extrinsics matrix.matrix. Ideally, over a sequence of images (planning to use the checkerboard utilities).

The receivers produce different image sizes, though. E.g. Cam1 is w_1 x h_1 and Cam2 is w_2 x h_2, where w_1 != w_2 and h_1 != h_2.

Given the description of the parameter, though:

imageSize – Size of the image used only to initialize intrinsic camera matrix.

This leads me to believe that it is assumed both images have the same dimensions. Given the goal of computation, I don't believe super-sampling one of the images to make them both the same size is a good decision.

So I should just be calibrating each camera individually?

Thanks for any suggestions!