johncostella

7 ms·

1.
by johncostella 2y ago
I'm biased, but Magic Kernel Sharp does a good job for many applications. :) https://johncostella.com/magic/
2.
by johncostella 5y ago
The kernels themselves should be applied in linear space, and their properties are independent of the signals to which they are (assumed linearly) applied. If the practical examples I applied them to should have had gamma compression remove
3.
by johncostella 5y ago
Yes, and if you're comfortable pulling down the code and running 'make' on any *nix system, there is an executable 'resize_image' that lets you test it out for yourself. I'm always looking out for corner cases
4.
by johncostella 5y ago
Thanks. It was unexpected to me as well. The original "magic" kernel was really along the lines of interpolation (particularly by large factors — there were all sorts of algorithms out there, e.g. repeatedly upsizing by small fact
5.
by johncostella 5y ago
Ah nice, I will correct that statement, which I was sure was bound to have a counterargument. :)
6.
by johncostella 5y ago
Nice — it did look to be a more general class of interpolating functions, and that does indeed look to be the case. Interesting!
7.
by johncostella 5y ago
Ah thanks, I wasn't aware of this. That mainly talks about the original "magic" kernel ([1,3,3,1]) and an earlier version of my page. It rightly references the fact that the magic kernel (any version) by itself slightly blurs
8.
by johncostella 5y ago
They are actually pretty good — I believe much better than truncated sinc (likely because of the discontinuity in first derivative of the latter), but I haven't done that analysis myself.
9.
by johncostella 5y ago
Granted, it's a bad example for a Security Engineer to not offer HTTPS. Unfortunately I use a provider that doesn't make it easy at all. I will get there eventually.
10.
by johncostella 5y ago
I discuss Charles Bloom's 2011 article in some detail on my page. The original "magic" kernel (i.e. [1,3,3,1]) can be rightly categorized as just many things. It's the generalization to the continuum case, and the additi
11.
by johncostella 5y ago
The paper includes examples of Magic Kernel Sharp 3 (the original), 4, 5, and 6. The C code allows you to use any value 'a' of these "generations," and any version 'v' of the Sharp kernel listed in the paper (u
12.
by johncostella 5y ago
I generated bicubic and Lanczos versions in 2006. It was stated that the Lanczos one evidently came from a version of GIMP that was buggy, so I removed it a long time back. The bicubic is still there from that era. If it's wrong, I wil
13.
by johncostella 5y ago
"Not obvious": yes, I was originally writing up the paper just to contain the derivation and results of the Fourier transform calculation — whatever it turned out to be — and was dumbstruck when it turned out to be sinc^3 x. The o
14.
by johncostella 5y ago
Right. Fundamentally, Magic Kernel Sharp seeks to be an efficient but good approximation to a rectangular low-pass filter; it doesn't add information, and ideally doesn't destroy or distort it either. Yes, I have seen other method
15.
by johncostella 5y ago
I put the Magic Kernel into the public domain in 2011, long before starting at Facebook. Yes, the Sharp step is nothing more than a three-tap sharpening filter, not at all remarkable in itself, and yes the "+" is just extra sharpe
16.
by johncostella 5y ago
Yes I have recently been asked about splines, and the Magic Kernels are of course a subset of those, but generated in a specific way. I have not had a chance to compare to all previous interpolation kernels (and there are many of them). Yes
17.
by johncostella 5y ago
As I just told a colleague who told me of this post, "I'm so old I don't even know how to use that [this] site." :) Happy to answer questions or address criticisms.