-
Notifications
You must be signed in to change notification settings - Fork 141
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[GetGlobalPlacement] Issue with the scale computation #159
Comments
Maybe a solution, given a |
I don't know how to do the polar decomposition yet. But I know for sure there is no way to distinguish between a matrix representing no scaling at all and one that is a combination of scale(-1,-1,1) and a rotation of 180 degrees. Only option is to assume that all scale factors have the same sign. There is work going on to fix related issues in the creation of a rotation from a matrix. Maybe following code will work for now.
|
The macro return the wrong result for an object with rotation
0.45642222410108496, 0.49809724456331667, 0.5849212871934979, -0.4488262203714842
. The returned scale factor is-1, -1, 1
but should be1, 1, 1
. According to this, the scale computation involves the eigenvalue decomposition.I already fixed the issue for objects that are not scaled in 47dbb41. The issue subsists for scaled objects but a warning is printed.
I added a test for this.
@Jolbas do you have the capacity to have a look at this?
The text was updated successfully, but these errors were encountered: