You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Thank you for all your work on this library! This is not a true issue, but rather a conceptual question -- I hope you don't mind.
I'm interested in a problem scenario where we don't expect arbitrary rotations. Are you aware of any PSR algorithms that allow limiting constraints or regularization on the rotation/scale? So far I'm mostly familiar with CPD which (if I understand correctly) does not easily support this because of the closed-form update equations. I tried to search for this myself but couldn't find anything explicitly addressing this capability.
If you could point me to a paper to start with that might address this, I'd be very grateful!
The text was updated successfully, but these errors were encountered:
Hi,
Thank you for contacting me!
I too am interested in registration methods with restricted rotation axis, but unfortunately I have not been able to do much research on the methods out there.
Is PSR the name of the algorithm you mentioned or something else?
If possible, could you please share the URL or something?
Thanks for your response! Yes I’ve searched a bit but haven’t had any luck finding existing methods that account for restricted rotation yet. I’d be very interested if you happen to come across anything. Or maybe it’s an opportunity for new research.
By “PSR”, I meant “point set registration.” Just another term for the general problem. The main technique I’ve used so far is coherent point drift.
Thank you for all your work on this library! This is not a true issue, but rather a conceptual question -- I hope you don't mind.
I'm interested in a problem scenario where we don't expect arbitrary rotations. Are you aware of any PSR algorithms that allow limiting constraints or regularization on the rotation/scale? So far I'm mostly familiar with CPD which (if I understand correctly) does not easily support this because of the closed-form update equations. I tried to search for this myself but couldn't find anything explicitly addressing this capability.
If you could point me to a paper to start with that might address this, I'd be very grateful!
The text was updated successfully, but these errors were encountered: