It seems that the calculation of rotation matrix have a mistake?

ClayFlannigan / icp

iterative closest point

Other

603 stars 153 forks source link

It seems that the calculation of rotation matrix have a mistake? #6

Closed 446810140 closed 6 years ago

446810140 commented 6 years ago

Dear Clay Flannigan, According to equation 16 in the paper "Least-squares Rigid Motion Using SVD" ( http://igl.ethz.ch/projects/ARAP/svd_rot.pdf ), it seems that H should be equal to np.dot(AA, BB.T) in icp.py?

446810140 commented 6 years ago

Hmm, I understand it. Matrix A and B have been transposed when input, so it is right. Thanks & Best Regards!