I have already get the intrinsic parameters by your toolbox. However I need to use another toolbox called 'OCamCalib' (https://sites.google.com/site/scarabotix/ocamcalib-toolbox) to get rectified images. I need the Taylor series parameters. Do you have any idea about how to change [xi fu fv pu pv] to [a0 a1 a2 a3 a4 a5 ...].
here is part of your toolbox parameters file:
distortion_model: radtan
intrinsics : [2.009638097118617, 935.8153370451462, 935.5369432030301, 366.0169636250961,242.43646884162092]
This is part of 'OCamCalib' toolbox parameters file:
#polynomial coefficients for the inverse mapping function
8 458.302420 268.674834 5.598408 27.061157 -13.483564 -38.424966 -21.788390 -4.099298
#affine parameters "c", "d", "e"
0.999890 0.000187 0.000406
I thought 'validator' can give me rectified image but I am wrong. So, can I get rectified images just from your tool box?
Tairui
I am experiencing the same problem. In particular, I have found a small example in aslam_cv_python/example/prototype.py, which is supposed to do what we need, however it does not work as the required class "NCameraPipeline" is missing. When I try to compile this cpp file "src/NCameras.cpp", it does not compile (some includes are missing from the folder). Did anyone succeed to make this small script run?
Best regards,
Dmytro