search

WolframRhodium / muvsfunc

Muonium's VapourSynth functions
75 stars 19 forks source link

Seesaw #8

Closed ghost closed 6 years ago

ghost commented 6 years ago

In function SeeSaw() line 2077: Szrp = Szp / pow(Sstr, 0.25) / pow((ssx + ssy) / 2, 0.5) Szrp was not used anywhere (It should be szp) Original avs code: Szp = Szp / pow(Sstr, 1.0/4.0) / pow((ssx+ssy)/2.0, 1.0/2.0) Also, peak was not used in main function, only sharpen2 needs it.

ghost commented 6 years ago

funcName is not defined in Compare2, so it will raise NameError if at least one of input is not a clip.

WolframRhodium commented 6 years ago

Fixed. Thank you.

ghost commented 6 years ago

Was actually testing it, and find a bug introduced in commit cd5090c The sharpen2 function definition is __SeeSaw_sharpen2, but it should be _SeeSaw_sharpen2 (NameError: name '_SeeSaw_sharpen2' is not defined)

WolframRhodium commented 6 years ago

Done. Thank you very much. Sorry for my carelessness.

ghost commented 6 years ago

Latest awarpsharp2 has a new parameter "cplace". But the default is "mpeg1" to keep consistency with previous versions. I think "mpeg2" would be a better choice when actually using it since it is more widely used. But if you don't feel like changing it, that's fine.

WolframRhodium commented 6 years ago

I have noticed the change. Currently, awarpsharp() in my scripts always functions on GRAY, so I think there's no need to change it.

Anyway, I see filtering chroma-subsampled clip by awarpsharp() as nearly unnecessary, because both the spatial resolution of chroma and the computational cost are low. I always prefer to perform it on YV24.

ghost commented 6 years ago

@WolframRhodium My bad

But typically I won't upsample the chroma to full resolution if they're subsampled. I will prefer to process it on GRAY individually.

WolframRhodium commented 6 years ago

Oh I mean something like fix chroma bleeding which does processes on GRAY. I confuse it with KNLMeansCL.