Thanks for your reply Howard! I made some comparasions with the IBK, and as you say they aren't matching completely. However, this will defiantly come in handy, thanks a lot!
Funny, I would think the use bkg luminance would be quite a simple operation, but I guess there most be a lot of things going on. Cheers! 2013/4/2 Howard Jones <[email protected]> > Here's one in progress I've been working on - not a perfect match but I've > been looking at the very same thing. > If someone has a better version and want to share or can improve on this ;) > > Howard > > set cut_paste_input [stack 0] > version 7.0 v4 > push $cut_paste_input > Group { > name screenSubtract > knobChanged "\nn=nuke.thisNode()\nk=nuke.thisKnob()\nif > k.name()=='screenOp':\n > op=n\[k.name()].value()\n > nuke.toNode('screenSubtractOp')\['operation'].setValue(op)\nelif > k.name()=='lumaOP':\n > op=n\[k.name()].value()\n > nuke.toNode('luminanceOp')\['operation'].setValue(op)\n \n" > selected true > xpos -179 > ypos 284 > addUserKnob {20 User l Screen} > addUserKnob {6 scSubtract l "screen subtraction" +STARTLINE} > scSubtract true > addUserKnob {6 bgChroma l "use bg chroma" -STARTLINE} > bgChroma true > addUserKnob {6 bgLuma l "use bg luminance" -STARTLINE} > bgLuma true > addUserKnob {26 ""} > addUserKnob {4 screenOp l "screen subtract op" M {difference from}} > addUserKnob {4 lumaOP l "luminance op" -STARTLINE M {geometric multiply > overlay ""}} > addUserKnob {26 ""} > addUserKnob {26 edgeAdjust l "<b>adjust edge"} > addUserKnob {7 edgeWP l "white point"} > edgeWP 1 > addUserKnob {26 ""} > addUserKnob {26 rmBGNoise l "<b>BG Noise"} > addUserKnob {7 BGNoiseThreshold l threshold R 0 0.3} > } > Input { > inputs 0 > name matte > xpos -318 > ypos 328 > } > Dot { > name Dot12 > label M > xpos -287 > ypos 368 > } > Expression { > channel3 alpha > expr3 a?a:0 > name addAlphaChannel > xpos -318 > ypos 417 > } > set N2cf09f10 [stack 0] > Dot { > name Dot15 > xpos -287 > ypos 1437 > } > set N2cf1d620 [stack 0] > Dot { > name Dot2 > xpos -287 > ypos 1714 > } > push $N2cf1d620 > Grade { > channels alpha > whitepoint {{edgeWP}} > black_clamp false > name Grade1 > xpos -479 > ypos 1419 > } > Input { > inputs 0 > name Bg > xpos -827 > ypos 392 > number 1 > } > Dot { > name Dot16 > label BG > xpos -796 > ypos 473 > } > Dot { > name Dot17 > xpos -796 > ypos 672 > } > Dot { > name Dot3 > xpos -749 > ypos 672 > } > set N2cf394b0 [stack 0] > Dot { > name Dot18 > xpos -528 > ypos 672 > } > Saturation { > saturation 0 > mode Maximum > name Saturation1 > xpos -559 > ypos 843 > } > Shuffle { > red blue > green blue > alpha blue > name Shuffle2 > label "\[value in] > \[value out]\n\[value in2] > \[value out2]" > xpos -559 > ypos 1079 > } > push $N2cf09f10 > Invert { > name Invert2 > xpos -414 > ypos 443 > } > Input { > inputs 0 > name CS > xpos -481 > ypos 327 > number 2 > } > Dot { > name Dot13 > label C > xpos -450 > ypos 368 > } > Copy { > inputs 2 > from0 rgba.alpha > to0 rgba.alpha > name Copy3 > xpos -481 > ypos 510 > } > Premult { > name Premult3 > xpos -481 > ypos 595 > } > push $N2cf09f10 > Input { > inputs 0 > name Fg > xpos -155 > ypos 332 > number 3 > } > Dot { > name Dot14 > label FG > xpos -124 > ypos 390 > } > Copy { > inputs 2 > from0 rgba.alpha > to0 rgba.alpha > name Copy2 > xpos -155 > ypos 450 > } > set N2cf78db0 [stack 0] > Merge2 { > inputs 2 > operation difference > also_merge all > name screenSubtractOp > xpos -155 > ypos 595 > } > push $N2cf78db0 > Dot { > name Dot5 > xpos 42 > ypos 516 > } > Premult { > name Premult1 > xpos 65 > ypos 641 > } > Switch { > inputs 2 > which {{scSubtract}} > name doScreenSubtraction > xpos -155 > ypos 734 > } > Dot { > name Dot19 > xpos -124 > ypos 788 > } > set N2cf9b720 [stack 0] > Dot { > name Dot20 > xpos -345 > ypos 788 > } > set N2cf9f060 [stack 0] > Colorspace { > primary_in "Wide Gamut" > colorspace_out CIE-Lab > primary_out "Wide Gamut" > name ColorspaceColor1 > xpos -690 > ypos 783 > } > push $N2cf394b0 > Colorspace { > primary_in "Wide Gamut" > colorspace_out CIE-Lab > primary_out "Wide Gamut" > name ColorspaceIn1 > xpos -780 > ypos 787 > } > ShuffleCopy { > inputs 2 > red red > alpha alpha2 > name LABshuffle1 > xpos -721 > ypos 850 > } > Colorspace { > colorspace_in CIE-Lab > primary_in "Wide Gamut" > primary_out "Wide Gamut" > name ColorspaceOut1 > xpos -721 > ypos 918 > } > push $N2cf9f060 > Switch { > inputs 2 > which {{bgChroma}} > name Switch1 > label "use bg chroma" > xpos -721 > ypos 1091 > } > set N2cfd5db0 [stack 0] > Merge2 { > inputs 2 > operation geometric > also_merge all > name luminanceOp > xpos -559 > ypos 1263 > } > push $N2cfd5db0 > Switch { > inputs 2 > which {{bgLuma}} > name Switch2 > label "use bg Luminance" > xpos -721 > ypos 1277 > } > push $N2cf9f060 > Dot { > name Dot21 > xpos -345 > ypos 1369 > } > Keymix { > inputs 3 > invertMask true > name Keymix2 > xpos -721 > ypos 1433 > } > push $N2cf9b720 > Dot { > name Dot1 > xpos -124 > ypos 1551 > } > Copy { > inputs 2 > from0 -rgba.alpha > to0 -rgba.alpha > channels rgb > name Copy4 > xpos -721 > ypos 1547 > } > Copy { > inputs 2 > from0 rgba.alpha > to0 rgba.alpha > name Copy1 > xpos -721 > ypos 1718 > } > Unpremult { > name Unpremult1 > xpos -721 > ypos 1826 > } > Grade { > channels alpha > blackpoint {{BGNoiseThreshold}} > name BGNoiseRemove > selected true > xpos -721 > ypos 1882 > } > Premult { > name Premult2 > xpos -721 > ypos 1937 > } > Output { > name Output1 > xpos -721 > ypos 2147 > } > end_group > > > > ------------------------------ > *From:* Simon Björk <[email protected]> > *To:* Nuke user discussion <[email protected]> > *Sent:* Tuesday, 2 April 2013, 22:14 > *Subject:* [Nuke-users] IBK and use bkg luminance > > I've been trying to understand how the IBKs "use bkg luminance" switch > works, as I've needed it for other keyers, but I haven't been able to come > up with a match. I can get somewhere close by converting the fg and bg to > HSL color-space, and combining the luminance channels, but I can't get a > perfect match compared to the IBK. Anyone know what's happening under the > hood? > > _______________________________________________ > Nuke-users mailing list > [email protected], http://forums.thefoundry.co.uk/ > http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users > > > _______________________________________________ > Nuke-users mailing list > [email protected], http://forums.thefoundry.co.uk/ > http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users >
_______________________________________________ Nuke-users mailing list [email protected], http://forums.thefoundry.co.uk/ http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users
