Enlarging Digital images

Ryadia@Home wrote:

Somehow I get the idea you are as big a fool as Gisle


I'm marking this thread IGNORED. This is a bunch of bull**** talk
wasting my time.

These links and references seem to back up my impression that QImage's
Vector method, alongside Genuine Fractals, used to be the leaders. (I
didn't particularly like GF when i tried it, as it has unpleasant
artefacts in some circumstances.) But now QImage's new Pyramid
interpolation appears to be well ahead of those two.

Unlike a certain person here, Mike Chaney from DDISoftware seems to
have a genuine, testable, provable, breakthrough in his new method. I
would invite anyone reading this thread to go and visit that dpreview
link (
http://forums.dpreview.com/forums/re...essage=7642362
), and compare Mike's attitude (and in return, the attitude of his
customers) with that of Douglas.

I think I might just pop over and buy QImage.. (yes, with my next
*dole cheque*, Douglas....)

(O;

There is a comprehensive comparison of all methods including QImage
Pyramid, Genuine Fractals and PS's bicubic smoothing he

http://www.charm.net/~mchaney/testpics/resample.jpg


Admittedly this is supplied by QImage's creator, but then I think he
has shown *he* can be trusted, and he *knows* his stuff, and does not
underestimate the comprehension of his audience...

Crownfield wrote:

Ryadia wrote:

Crownfield wrote:



http://vircen.com/2-pixel/2-pixel-kitty.jpg

alas, not a redhead, but a pretty kitty.
I also have the image with all the original 6 mp of detail.
you show me yours, i'll show you mine.


Douglas

Sorry crownfield...
6x4 pixels is too small to sample
and it's not the 1 pixel image you told me work on...

i was generous.

you were going to demonstrate the power of your software
on one pixel, and I gave you many more than that.
It should be a cinch.

since you know it is a kitty, that should help your software to
intelligently create the missing detail.

Tomorrow, I will show you what your software could have generated.

http://vircen.com/pixel/DSCF0025.JPG

In message ,
"Ryadia@TA" wrote:

Your understanding (and a few others who can't grasp the finite size of
a digital image) of what constitutes detail and my practicing use of the
description differ... That's all.

Why can't you just state it as it is:

It's not detail that is being added, but rather, *TEXTURE*; texture that
satisfies the eyes and brain that no further focusing is necessary.


--


John P Sheehy

Sorry for continuing this ridiculous thread, but...

Tomorrow, I will show you what your software could have generated.
http://vircen.com/pixel/DSCF0025.JPG

Yep, that's pretty well *exactly* what I got with my Very Speshul=99
interpolation program. I'm not going to post the result, because you
guys might reverse engineer it. Just trust me, ok? And besides, it
looks just like yours anyway, so what is the point?

Dunno what's so hard about all this. Clearly, all digital images are
infinitely enlargable. (O:

Chrlz wrote:

Dunno what's so hard about all this. Clearly, all digital images are
infinitely enlargable. (O:

I've written a compression algorithm that will compress a file of any
size down to 1 bit. I confess that the decompression agorithm is still
not completed. Having a spot of trouble there...

Cheers,
Alan



--
-- r.p.e.35mm user resource: http://www.aliasimages.com/rpe35mmur.htm
-- r.p.d.slr-systems: http://www.aliasimages.com/rpdslrsysur.htm
-- [SI] gallery & rulz: http://www.pbase.com/shootin
-- e-meil: Remove FreeLunch.

Alan Browne wrote:
Chrlz wrote:

Dunno what's so hard about all this. Clearly, all digital images
are
infinitely enlargable. (O:

I've written a compression algorithm that will compress a file of
any
size down to 1 bit. I confess that the decompression agorithm is
still not completed. Having a spot of trouble there...


I understand some cutting-edge programmers are into negative
bit-space. I think they base the concept on some large reservoir
within their heads.

Only problem is the echos e c h o s e c h o s

In message ,
Alan Browne wrote:

I've written a compression algorithm that will compress a file of any
size down to 1 bit. I confess that the decompression agorithm is still
not completed. Having a spot of trouble there...

You can include metadata that has an image number, and the image can be
downloaded from a library based on this number. If metadata size is not
an issue, you can use a number that defines the image without a library.
--


John P Sheehy