Deform curves with other curves in loop

[sjrohit]

hey Guys,

I am trying to wrap bunch of head hair guide curves with my simmed hair curves which have a different topology.  I had to resample the curves during simulation to achieve a better sim. If I just do a basic point deform setup, it produces some artifacts because the curves are in close proximity of each other. I tried using "for each primitive" to run it over each curve primitive but its still not giving in accurate results. Any idea if there is a better way to do this?

Appreciate your help!

thanks!

[animatrix]

Hi,

You can capture point neighbours and weights inside a For Loop network using Point Deform (Capture), and then use another Point Deform SOP to deform afterwards.

So the capturing is done without time dependency which takes longer to cook while the deformation happens after, which will run much faster.

[sjrohit]

Thanks @animatrix,

Thats a good idea for saving some cooking time, I will try that.

But the real problem I am having is to setup the point deform inside for loop, its creating stretch in the curves which should not happen if I am wrapping the curves 1 to 1. I am sure I am doing something wrong. Not sure if I should keep the timeshift nodes inside the For Loop and there is only one input on ForeEach node, If at all I need to keep the timeshift inside For Loop, how do I create multiple inputs?

Attaching an example hip file, if you don't mind taking a look, I would highly appreciate.

Its a basic vellum sim done on curves and I am trying to wrap it to the curves on right side (different topology). If you go to frame 70, You will see some artifacts in the curves. I am sure my for each setup is not quite right.

 

thanks!

 

 

curve_wrap.hipnc

[Aizatulin]

Hi,

you can also try wire deform for this purpose.

curve_wrap_wiredeform.hipnc

[sjrohit]

thanks Aizatulin,

I saw your file, but there is an offset between the simmed curves and wrapped curves, I am looking for a one-to-one wrap where the wrapped curves follows the simmed curves almost perfectly. 
 

[Aizatulin]

I'm expecting, that there will be an offset since there are fewer guide curves, but it would be nice if you can provide a simple picture (or sequence of picutres) of your final goal.

[sjrohit]

Sure thing!

Actually the number of guide curves are same in simmed and render geo. I have just resampled the curves in render geo to give it a slightly different topology. Will provide some images, thank you