[ptx] Optimizations

[David Grant]

David Grant wrote:

> Which optimizations do people normally do in practice? One of the 
> tutorials advises not to use "everything" but perhaps just use the 
> first two, simplest, optimizations.
>
Here's the quote from Rob Park: "It might be tempting to optimize for 
"Everything", but that is recommended only when there is 50% overlap, 
and the control points are placed over the whole overlap area. Otherwise 
the optimizer reduces control point error close to the boundary, but add 
strange bubbles or other distortions in the middle."

But on the Hugin tutorial by Ed Halley: "Once you get good results, you 
can get better results by optimizing more variables. You can optimize 
for every variable at once, but this is slower, and can even fail if 
your initial guesses weren't good enough."

And on Bruno's tutorial:

"The lens parameters can be optimized at the same time; for now we can 
leave the *HFOV* (horizontal field of view) unaltered, but can try and 
correct barrel distortion by selecting the *b* parameter.

/Tip: Although you can optimize all hfov, a, b, c, d & e parameters, 
most of the time the hfov and b parameters are the only significant 
numbers you need to consider."/

It seem like everyone has their own preference.

-- 
David J. Grant
http://www.davidgrant.ca:81

-------------- next part --------------
A non-text attachment was scrubbed...
Name: david.grant.vcf
Type: text/x-vcard
Size: 180 bytes
Desc: not available
Url : http://www.email-lists.org/pipermail/ptx/attachments/20050106/0f082baa/david.grant.vcf