Geek asks and answers

OpenCV core and essential matrix (python) disagree

I am trying to calibrate a stereo camera using OpenCV (python interface). I first calibrated the two cameras separately using calibrateCamera2, and then passed the parameters to stereo calibration

cv.StereoCalibrate(object_points, image_points_left, image_points_right, \
               point_counts, intrinsic_left, distortion_left,\
               intrinsic_right, distortion_right, \
               (IMGRES_X,IMGRES_Y), R, T, E, F, \
               term_crit=(cv.CV_TERMCRIT_ITER+cv.CV_TERMCRIT_EPS, 100, 1e-8),\
               flags=cv.CV_CALIB_FIX_INTRINSIC)

I check the result with an epipolar constraint (as described in the OpenCV book) and get an average error of around 0.0039.

In principle, I should be able to connect the fundamental and essential matrix with the matrices of my camera. So what I do:

Mr = asarray(intrinsic_right,dtype=float64)
Ml = asarray(intrinsic_left,dtype=float64)
E = asarray(E)
F = asarray(F)
F2 = dot(dot(inv(Mr).T,E),inv(Ml))

However, the resulting matrix F2 is completely different from F. Is there something obvious that I'm doing wrong? Help is much appreciated.

Edit: dot and inv from numpy.

+3
source share
2 answers

E F, StereoCalibrate(), . F , , F F-, E, , , . , , . StereoCalibrate() F, F2, . , , .

+4

Matix Song...

, , - "" ? numpy dot, -, .

, :

A = mat(random.rand(3,3))
B = mat(random.rand(3,3))
dot(A,B) == A*B

, :

F2 = np.linalg.inv(Mr.T) * E * np.linalg.inv(Ml)

(N.B. numpy )

0

More articles:


All Articles