Mon, Nov 11, 3:41 AM CST

Renderosity Forums / Poser - OFFICIAL



Welcome to the Poser - OFFICIAL Forum

Forum Coordinators: RedPhantom

Poser - OFFICIAL F.A.Q (Last Updated: 2024 Nov 11 2:56 am)



Subject: Group tool, morph targets, and the graph... need help, please


Meddlesom ( ) posted Mon, 16 June 2003 at 12:45 PM · edited Mon, 04 November 2024 at 7:16 AM

Hopefully I can explain this well enough to be understood. What I am trying to accomplish is this, morph a body part, but I would like to select certain polygons in that body part that I can dial down on the morph. Example, say I have a full head morph for victoria that has elf ears and I would like to be able to unmorph only the ears to make her human again. Is it possible to select the polygons of the ears and use the graph to adjust the dial value of just the ears? If that is not possible, is there another solution? I've looked through the forums and haven't really found anything about this, and have gone through several group tool and morphing tutorials that don't really give any answers either. I am having great difficulty with this issue... any help would be greatly appreciated. Thanks in advance!


ockham ( ) posted Mon, 16 June 2003 at 1:37 PM

If there isn't a more obvious method, I think this should be possible with Python. The script would have to create a new separate morph target that excludes the facets contained in any Groups that are active. The opposite should also be possible: morph -only- what's contained in Groups. Let me know if you haven't found any better method, and I'll try it.

My python page
My ShareCG freebies


Meddlesom ( ) posted Mon, 16 June 2003 at 4:46 PM

That sounds like a good idea, unfortunately I have no knowledge of python scripting. =(


ockham ( ) posted Mon, 16 June 2003 at 5:00 PM

I'll give it a try. Should be interesting. If you have either ProPack or P5, you can use the script without knowing anything special!

My python page
My ShareCG freebies


Meddlesom ( ) posted Mon, 16 June 2003 at 9:13 PM

I appreciate you trying, and I'm sure a script like this would be helpful to a lot of people.


ockham ( ) posted Mon, 16 June 2003 at 10:48 PM

I wish I had made this one a long time ago! I've always needed an easy way to make one-sided morphs, and this does it. Pick up the ZIP at the URL given in next message, and unzip into the same folder with POSER.EXE as usual. It will create a separate folder under Runtime:Python, and put the scripts and the Readme into that folder. You can then run either of the two scripts. The Poser manual discusses running scripts, or you can look at my one-picture tutorial at ockhamsbungalow.com/Python/manual.jpg See next few msgs for an intro.

My python page
My ShareCG freebies


ockham ( ) posted Mon, 16 June 2003 at 10:50 PM

Attached Link: http://ockhamsbungalow.com/Python/MTbyGroup.zip

file_63037.jpg

Start by selecting the desired body part and forming your groups. (The URL on this message is the ZIP)

My python page
My ShareCG freebies


ockham ( ) posted Mon, 16 June 2003 at 10:52 PM

file_63038.jpg

Activate one of the two scripts. In this case I've used MTincGroup, which forms up the morph using only the facets inside the group.

My python page
My ShareCG freebies


ockham ( ) posted Mon, 16 June 2003 at 10:54 PM

file_63039.jpg

And here's the result: a one-sided smile using only the facets inside the red area. Note the new MT dial (which isn't easy to read) says SmileINCleft, using the original morph name, plus INC for include, plus the name of the first group. Try it out and see how it works. Tell me about any problems!!!!!

My python page
My ShareCG freebies


Migal ( ) posted Tue, 17 June 2003 at 6:45 AM

This is interesting stuff, Ockham.

If it could generate deltas from selected polygons via the trans, scale and rotate, you'd have an original morph creation system inside Poser, and it might be easier to control than magnets.

Nice work!


ockham ( ) posted Tue, 17 June 2003 at 10:13 AM

Hmm.... Just thinking about the face morphs I've tried to create in Amapi (with far too much work)..... Seems like the relevant directions would be 1. Bulge out or in from center of head. (Move along the normals.) 2. Rotate the group around the center of the head, in any of the three axes. These would have to be controlled by sliders on the Python panel. Some kind of smoothing around edges should be optional, maybe with a choice of how many facets to interpolate. Would these 4 directions work for other body parts as well, or is there another direction?

My python page
My ShareCG freebies


Migal ( ) posted Tue, 17 June 2003 at 1:30 PM

You'd want to be able to rotate around the center of the selected area, if possible. That would give a sharper angle. Translate would be the same thing as moving just the selected area in a planar fashion. For instance, to make the nose longer, you'd select just the bridge (leading edge) of the nose, then apply negative z trans to the selection. You're exactly right about move along normals. That would give you a 3D expand/shrink. Are you thinking it's possible?


ockham ( ) posted Tue, 17 June 2003 at 2:05 PM

Certainly possible. Now that I've figured out how to distinguish grouped from ungrouped, moving one set is fairly easy. Rotating around center may not be easy, or at least I don't immediately see how to do it. I see your distinction between Z-trans and bulge. That helps to clarify things.

My python page
My ShareCG freebies


Migal ( ) posted Tue, 17 June 2003 at 5:16 PM

Wow. If you could pull it off, I think it'd be a pretty big deal. Even if it could only be done for the head. Doing it to selected polygons across welded actors would be much more complicated, but required for the body.


ockham ( ) posted Tue, 17 June 2003 at 5:49 PM

"Doing it to selected polygons across welded actors would be much more complicated, but required for the body." I'm about 3/4 done with the basics, and it's surprisingly straightforward. Working on several actors at once -wouldn't- be any harder. If you set up all desired groups in advance, the script can act on all groups at once regardless of body part. What might get tricky in this situation is symmetry...... when to make one group move oppositely to another.

My python page
My ShareCG freebies


Migal ( ) posted Tue, 17 June 2003 at 6:22 PM

On a human figure, wouldn't that be inversion of the X delta coordinates? This sounds neat, Ockham. It might also be a quick way to fix poke-through on clothes.


Little_Dragon ( ) posted Tue, 17 June 2003 at 8:07 PM

I thought that P5's morph putty feature was supposed to work like this, but I still haven't figured it out.



Meddlesom ( ) posted Tue, 17 June 2003 at 10:21 PM

Thanks, ockham! This solves my problem perfectly, well done!


ockham ( ) posted Wed, 18 June 2003 at 11:10 PM

You're right, there is a similarity. I had to go read the P5 help file to see what it said; I remembered trying the Morph Putty thing once and finding it impenetrable. But (I think!!!!) Migal's idea is the converse of Morph Putty. Putty takes existing morph targets and modifies them; you can't affect anything that isn't already part of a set of deltas. Migal's idea is to start with your own set of chosen polygons, manipulate them, and spawn a morph target from the moved result. If somebody actually understands the putty thing, please chime in and clarify!!!!!

My python page
My ShareCG freebies


Migal ( ) posted Thu, 19 June 2003 at 3:42 PM

I understand a python tool with sliders would be significantly easier and probably more powerful. :-)


Privacy Notice

This site uses cookies to deliver the best experience. Our own cookies make user accounts and other features possible. Third-party cookies are used to display relevant ads and to analyze how Renderosity is used. By using our site, you acknowledge that you have read and understood our Terms of Service, including our Cookie Policy and our Privacy Policy.