It is quite intesting !!
RMS might be the simplest way for evaluating an error,
in the case that the algorithm uses a least square
method. Why don't you use angles between reference
axes?
----- Original Message -----
From: sman
To: rp-ml@bart.lpt.fi
Sent: Wednesday, April 05, 2000 8:30 AM
Subject: How to evaluate the registration error??
Hello!!
Have you integated the multiple sets of range images
into a common reference frame?
It's a registration which is essential process of
reverse engineering.
Dear sman.
During integrating the several data, how can you
evaluate the error?
As I know, RMS error is widely used. Do U have any
other ideas about that??
If U are interested in registration, I wish you share
the information with me.
You can say that every information, if it is related
to registration, will be helpful to me..
thanks for your concern.
__________________________________________________
Do You Yahoo!?
Talk to your friends online with Yahoo! Messenger.
http://im.yahoo.com
For more information about the rp-ml, see http://ltk.hut.fi/rp-ml/
This archive was generated by hypermail 2.1.2 : Tue Jun 05 2001 - 23:03:13 EEST