Open dstansby opened 7 months ago
Although I agree docs should be added, the form of the matrix is a 3x4 matrix which implies it’s a standard form where you have an xyz1 vector, and you multiply it into the 4x4 and read out the first 3 components.
this puts translation “after” rotation, in the sense that it is expressed in the destination frame.
It's currently possible to specify a coordinate transform for images, but it's not clear whether the rotation matrix is applied before or after the translation. It would be great to add a short section to the markdown docs on which order the rotation/translation happen in.