Matrix to quaternion

[HTLife]

I found a possible conversion bug that caused by this line.

Should the w(real) be M[0, 2]-M[2, 0]?

I use this website for verification.

[ -0.4285714,  0.2857143,  0.8571429;
   0.8571429,  0.4285714,  0.2857143;
  -0.2857143,  0.8571429, -0.4285714 ]

w[ 0.3780]
x [ 0.3780]
y [ 0.7559]
z [ 0.3780]

Matlab rotm2quat gave the same answer. (positive w)

If I use M[2, 0]-M[0, 2], the w(real) term will be negative.

[KieranWynn]

Hi HTLife

Thanks for the feedback. I can't reproduce this behaviour with pyquaternion. The matrix you provide above is not quite precise enough to satisfy pyquaternion's requirements for orthogonality, but using a very similar one derived from your desired output, I have the following:

R
array([[-0.42848504,  0.28568188,  0.85719685],
       [ 0.85719685,  0.42848504,  0.28568188],
       [-0.28568188,  0.85719685, -0.42848504]])
>>> q = pq.Quaternion(matrix=R)
>>> q
Quaternion(0.377993043102062, 0.377993043102062, 0.75588608804457313, 0.377993043102062)
>>> R2 = q.rotation_matrix
>>> R2
array([[-0.42848504,  0.28568188,  0.85719685],
       [ 0.85719685,  0.42848504,  0.28568188],
       [-0.28568188,  0.85719685, -0.42848504]])
>>> q.real
0.377993043102062
>>> q.imaginary
array([ 0.37799304,  0.75588609,  0.37799304])
>>>

Note that q is in the form [w, x, y, z]

I get a positive w in my output. This seems to match the output of the links you provided when run on the same matrix R.

Could you please provide some more concrete input or the exact matrix you used that will allow me to reproduce this?

Thanks

[subTropic]

Hi, I am also having an issue with the matrix to quaternion method, it looks the values it produces has the sign flipped for the i, j, k components.

I have verified this against quaternions produced by the Maya python 2.0 API, and the Code example on Euclidean space: http://www.euclideanspace.com/maths/geometry/rotations/conversions/matrixToQuaternion/

for example this 4x4 matrix (columns first): [0.056706403022355786, 0.00266710458741429, 0.9983872097054282, 0.0, -0.9903460785750324, 0.12683943299448722, 0.05591084229349291, 0.0, -0.12648577086793414, -0.9919195438539943, 0.009833966390741607, 0.0, 49.89087677001953, 40.20277786254883, 99.87789154052734, 1.0]

produces this with pyquaternion: i,j,k,w = (-0.4795916919681297, 0.51485406716052, -0.4545018725888551, 0.5462096214841845) maya produces this: i,j,k,w =[0.4795917496422797, -0.5148541248130897, 0.454501952810933, 0.5462096289237796] euclidean space code produces this: i,j,k,w =(0.4795916919681298, -0.5148540671605201, 0.4545018725888552, 0.5462096214841845)

[ghost]

Hi,

Just wondering if there is any update on this issue? THanks.