Yet again me.. :/ A side effect of my research on "genesisify"-ing non-genesis figures is the cloth adoption problem. I made a script to push figure morph into clothes, directly from poser, without the roundtrip of creating a new cr2 outside with an external tool or such. its just start script, select figure morph and its pushed into the clothes figure is wearing. conforming too. i'd welcome if someone would try it out and see if some cloths performs particulary well or bad, give an opinion about whether its useful or not. i think the existing best-in-class cloth morphers are still better. but if i could refine the technique to match their level is could be really useful and free or at least low cost. unfortunately my time is very limited so i cannot do much research personally. but if it would be judged useful i'd upload it somewhere

I would be glad to test the script.

I will see what I can do with it. Sounds very usefull.

If you're still looking for testers then this is something I'd be keen to look at.

I certainly want to help testing.
Sounds like a very good idea.

please send mail to mail AT colorcurvature dot com

Finally!!! Been waiting on this for years..

An earlier post made by me (now deleted) assumed you were talking about the m4/k4 project.

That said, I now have the new script and have been working with it.

You state there is more work to be done. Where? I'm getting morphs that I need moved into clothing items quite effectively?

If you could outline your concerns and known issues, I'd be happy to look for them, but this seems to be a working product!

TAKE MY MONEY DAMMIT!!!

Looking forward to this.

E-mail sent. :D

Thank you for the opportunity to test this.

  I have tested it on a few figures and it works really well. There is some distortion in the hips on skirts that I don't see using MC.  But clothing fits much better  than when using MC on default settings.   Also a suggestion would be to have a distance offset such as the normal multiplier in your PoseMorphRefitter, to counteract minor poke-through in tight fitting clothing. 

Overall I'd say this is a must have, and I'd recommend releasing it, as it is definately useful and a real time saver.    

thanks for the feedback. i am trying to make a ui for it currently, so that one can process multiple morphs ( or mainly, all that are actually in use )

skirts are a problem yes. they need special treatment. but maybe its as easy as ignoring the legs for the fitting and only fit against hip. i didnt try it yet.

Always more interested.

Adding the distance control parameter now.

For the record, I think there is a bug in the current version that will drag the cloth a bit towards the figure. That was for the K4 morpher... but normally it shouldnt scale. That might be the reason for some of the pokes you saw.

Ok, and a mass processing mode to push into all clothing that the figure is wearing. Thats less to click.

Given that all my latest figures are just a single FBM "away" from V3 or D3, this should be tremendous useful.

Please let me know where I can buy a copy once it's ready !

@Joe: if you want to try  mail at colorcurvature com

I hope the essential parts are stable. Maybe a falloff for distant vertices would be good to avoid hard breaks. Issues will come up anyway, software is never bug free, but the quicker we identify the issues the better...

Maybe also a force overwrite mode would be nice, in case one wants to try different settings and chose the best without having to reload and reconform the clothes

Is this anything like the shit system we have in DS, if it is then I hope you are happy with "sprayed on" clothing, because that is all your going to get with it.

some issues with complex erc, where the fbm dial controls other masters which control various hidden minimorph sets. thats a todo, but not today.

bejay: what is a spray cloth? i havnt used daz long time. you mean a drawn cloth that technically is a texture?

I think he means clothing vertex pushed left-right and up-down, instead of only in-out.

That is how I would read "sprayed". (Personal opinion here)

Looks like someone's not happy with the Genesis' Autofit tool. 

 ;-)

Of course if you convert from a skinny man to a buxom beauty or a bodybuilder, loose fitting cloth will look skin tight because mainly the "bumps" get pushed into the clothing.

The solution would be to first create a proxy morph that has the same volume as the original morph, but has a smoother shape. 

Automated cloth fitting is always a compromise.

If you want 100% perfect fitting clothes, you need to convert manually in ZBrush or another modeller.

"Sprayed on" is what you get from auto generated morph systems, they quite literally add the exact shape of the figures morph to the clothing, leaving you with clothing that, when morphed to match the figure, is so skin tight that it looks like it's been spray painted onto the figure like a second skin.

We've had them since DS3 and the morphs you get from them really aren't worth using.

Wrong tool Joe, auto-fit is a clothing converter, and it's no better or worse than any of the others, as I still have to clean up the converted mesh in my modeller no matter which one I use.

Have you tested this on the OSX version of Poser 2012?

OSX and Poser 9 would be very welcome.

Still to be clarified: ignore JCMs, or try to process them.

If yes, how to identify a particular JCM. A rotation dial can control multiple JCMs.

Hard to judge which local morph targets must fall in which group. One could try to group BY-NAME or BY-ERC code. And still, how to slave the new morph accordingly to the original JCM, that is not trivial. 

Maybe the cloth makers have experience on how to hande JCMs in a conform cloth?

i did a quick test this morning with a custom morph & it transferred great, both into a piece of clothing & into a conforming necklace i made.  thank you for letting me test this wonderful tool!

Me too!!! 

I am new to poser still, and the clothing has been kicking my a**. 

I bought quite a few outfits here at the Market, but haven't really been able to use them.  :(

Ok, I think its getting better.

I now group morph targets by  master param, morph name, and ERC signature.

If this combination is unique, I consider that a group of morph targets belonging together. Otherwise, if there are different master parameters and/or different ERC formulars controlling morph targets with the same name, I reject processing them.

I modified the UI more to display if there is confusing about a morph target. i

Whaw, looking good... :-)

 It cannot yet conform morphs that are not DELTA ADD 1.0 controlling their local targets.

Thats for JCMs. I'll try to conform JCMs too and depending on success it will work or not.

But basically I hope its done :-).

Ok. It seems one can copy the ValueOp from a template.

I'm done, I pushed my first JCM into a cloth.

So its complete now. Lets find the final bugs and call it done.

@Joe: Wanna join in?

Waiting for my test copy with baited breath!

Got it. Spent all morning testing.

AMAZING!

I haven't been able to make it fail yet! (And believe me... I've been using some pretty extreme morphs to test it!)

However, I did get a strange distortion in the hip area of this item, the top from the PoserWorld Rebecca outfit.  This with Blackhearted's GND FBM.  There are no bones below the hip; the lower skirt section is fitted with adjustment morphs.  It was converted with V4 in the Zeroed position with all of the clothing conformed and magnetized.  I tried ignoring the legs, but it made it worse.

The elbow poke is likely due to the JCM...

I'll send you v0.8 maybe that makes things a bit better.

Otherwise: Its always happening with GND FBM or its specific to this cloth?

It might relate to the settings used or the version fails to collect the active GND targets to recognize they belong together.

If you increase the value in the "do not move vertices..." that things that defaults to 0.02, does it get more "consistent"? Either consistently good or bad. The hard break might indicate that vertices nearer than 0.02 are being projected and vertices > 0.02 are being ignored. The distortion horizon could be vertices near 0.02

Skirts are catastrophy anyways :)

It's that top, Hannes, as shorter ones work well, although one jacket with buttons ends up with the buttons getting larger towards the bottom...

I'll keep my eyes open for the new version, but I need to go plow my driveway...

PS I tried increasing Do Not Move Verts from 0.02 to 0.3 (yes, 0.3) and it didn't make much difference.

here i added another bulge to the figure (the poke). re-pushing Custom_Morph into cloth fixes the poke

p.s. best if you can find bugs with freebie items that i can also download :)

i'll have to browse some sharecg i guess.

Is this figure specific, or will it work with any figure?

Basically any, but quality highly depends. I cannot handle dresses well at the moment.

Buttocks area also, its the same kind of problem. GND4 Morph changes the figure lots in the hip area. That makes distortions occur in the morph.

Ah well, I do not know if I can fix that easily. I need a technical way of pretending the gap between left and right leg didnt exist at all. Fill the gap somehow, with dummy geometry or such, something that is easier to match the cloth against.

Instead of making it fully automatic how about semi-automatic?

Make a morphable leg filler prop, that the user of the program has to position and has to fiddle with until the figure has no legs, like a mermaid.

This sounds like an extremely good solution to me! Will it work CC?

The problem is allways between the legs.

Blender also has a problem between the legs for the Shrinking of clothing.

Build a helper object.

Save out the figure and the helper object as a single object.
Do your shrinking to this single object.

I build the helper object in this demo to knee height,  to show the result above and below the knee.

The shape and position of the helper object is not critical and the same helper object can be used for allmost all figures. => Because most figures are similar as I showed above.

Above the knee the clothing shrinks normally
And below the knee (where there is no helper object, the deformations .

Hope this can be of some help.
Happy Posering
Best regards
Tony

Thanks for you comments. I am however not sure if it could work that way, because a prop wouldnt morph with the figure carrying the cloth.

I have thought, maybe I can compute two or three vertex positions (one projection on left leg, one projection on right leg, one projection on hip) and make a weighted sum of these depending on the distance.  Also I have some idea for the calculation itself that I want to test.

Will keep you informed, but it will take a while. This week and the next will be busy.