Re: interpolation for a color image?

AE lover wrote:

If so, why don't we take into account the interaction between three
channels?

What interaction did you have in mind?

R1 is a convolution over the spectrum of color 1, R2 is the same convolution over the spectrum of color 2 ... if I make a weighted blend of color spectrum 1 and 2 and apply the convolutions, hey what do you know I get the same result as blending R1 and R2 with the same weights.

I see no interaction with the other colors. Of course this is assuming linear RGB ... it's non trivial to get close to linear RGB, but black level correction and gamma expansion will probably get you close enough. Especially since you are using bilinear interpolation :)

Marco
.

Relevant Pages

  • Re: Artifact Audibility Comparisons
    ... complex frequency spectrum of the music signal. ... domain ==> convolution of the complex spectra in the frequency domain. ...
    (rec.audio.pro)
  • Re: Artifact Audibility Comparisons
    ... complex frequency spectrum of the music signal. ... domain ==> convolution of the complex spectra in the frequency domain. ...
    (rec.audio.pro)
  • Re: Visualizations of common Newbie problems
    ... and convolution. ... On the left side draw a notional time signal. ... On the right side draw the notional spectrum. ... adding zeros in time increases the sample rate without changing the ...
    (comp.dsp)
  • Re: band-pass FIR filter low stopband attenuation problem
    ... >>> stopband attenuation, on only one side. ... >> convolution should be in time. ... >> I've never heard of inverting the spectrum in order to design a filter. ... >> Why not use the windowing method or Parks-McClellan? ...
    (comp.dsp)
Loading