AngleAxisToRotationMatrix and AngleAxisRotatePoint give different results?

[Mike Wu]

I have the following code

int main(int argc, char **argv) {

    double angle = 60 * M_PI / 180;

    double axis[3] = {0, 0, 1 / angle};

    double r[9];

    ceres::AngleAxisToRotationMatrix(axis, r);

    double pt[3] = {1,1,1};

    double pt_rotate[3];

    ceres::AngleAxisRotatePoint(axis, pt, pt_rotate);

    double pt_rotate_matrix[3];

    for (int i = 0; i < 3; ++i) {

        pt_rotate_matrix[i] = pt[0] * r[i * 3] + pt[1] * r[i * 3 + 1] + pt[2] * r[i * 3 + 2];

        LOG(INFO) << "diff i: " << pt_rotate_matrix[i] - pt_rotate[i];

    }

}

and they output 

I0514 17:21:13.835876  6115 localizer.cpp:230] diff i: 1.63255

I0514 17:21:13.836115  6115 localizer.cpp:230] diff i: -1.63255

I0514 17:21:13.836140  6115 localizer.cpp:230] diff i: 0

Is this a bug or I'm not using the functions correctly?

[Sameer Agarwal]

The rotation matrix is column major, you are interpreting it as row-major.

Sameer

--
You received this message because you are subscribed to the Google Groups "Ceres Solver" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ceres-solver...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ceres-solver/4ac44c3e-c71c-421d-bb04-89f5537c74ce%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Mike Wu]

Ah that explains it. Thanks for the reply. BTW is it documented anywhere?

[Sameer Agarwal]

It is documented in the header file, but the docs can be better.

To view this discussion on the web visit https://groups.google.com/d/msgid/ceres-solver/7978bbe4-18e4-43d7-994d-111ed569048b%40googlegroups.com.