1 | initial version |
The extrinsic parameters describe the rigid motion from the points of your object coordinate system to your camera coordinate system. If you use the model of Zhang, you can use nonlinear optimization to get the four intrinsic parameters. Are your images distorted? Then you should also consider them...
Another way would be to use the calibration-functions of openCV. Is it a problem for you if openCV calculates both parameter sets? You can just use the intrinsic parameters or compare the extrinsics of openCV against your data... so what's the problem? ;-)