So some of you helped me out when I was texturing a self portrait I was doing a few weeks ago. I had to get it all finished for an art show, but now I want to put it onto a body. I could create sort of a base with 3DS Max, but I’d rather work with ZSpheres. The problem is, when I use ZSpheres, there’s not really a hole or a spot to really connect the body to the head or at least cover it very well with the type of clothing I want my character to have. I’m still new to ZBrush, and I’m not really sure how to add or subtract geometry in any other way but using the brushes, and I know they’re not always the best solution. So would it just be better to create something in 3DS Max? Or is there a good way to get my head onto my body using ZBrush and ZSpheres?
There’s a very neat way to do this in zbrush, but it takes some time wrapping your head around.
- Have your head and body ready as seperated subtols.
- Place the head on top of the body roughly where they would connect.
- Duplicate the body and the head.
- Zremesh the two copies of the head and the body for nice topology as low as possible. (4000 faces)
- Merge the two zremeshed subtools together.
- Make a clone of the merged subtools.
Now you should have.
•Your two original high res models.
•Two zremeshed versions of your high res models. (Merged)
•A clone of your zremeshed model in the tool menu.
Now you have to.
- Insert a zsphere, shrink it a bit and put it somewhere where it won’t be in the way, but keep it at the symmetry line.
- Go down to topology and click “Select topo”, select the cloned zremeshed mesh you made.
- Now click edit topology and you will be able to delete and add vertexes/faces and they’ll wrap around the zremeshed subtool you’ve got. (You delete vertexes by holding down alt and clicking at the vertexes and you add new vertexes by clicking on the surface of the model, ctrl click a vertex to work from that vertex.)
- Now connect the zremeshed subtool at the neck when you’re done click “Adaptive skin” set the density to 1 and make an adaptive skin.
- Replace the old zremeshed model with your new edited topology model, Insert/your new adaptive skin model.
- Project back the detail of the two original head and torso models you had onto your connected topology.
You’ll have to smooth out the crease that will appear at the neck, now your two subtools will be connected. This may seem quite complicated but you really just got to do it a few times to understand it. If you don’t want to do this inside of zbrush you could just as easily export the zremeshed subtool and connect them in another program and then reimport them again. But if you understand how to do it in zbrush it’s probably faster to do it this way.
Djordr, would give a different result if I dynamesh a copy both parts together and then zremesher and then project detail? It could avoid the step of editing the topology.
You could do that but the topology won’t be too good and you would have to pull the dynamesh resolution up high to keep the detail around fingers, ears etc. Having good topology saves you a lot of heartache, at some point you would want to reduce that 15 million polygon model down to something that can be textured. But yes if you’ve just started out making a model dynamesh is useful but otherwise I would go with topology editing.
When you understand how to edit topology in zbrush it’s very fast, merging a torso to a head would take me probably around 2 min and there would be no loss of detail or any odd artifacts around fingers, lips or any other areas that you don’t want to have merged. The crude way would be to use dynamesh, it works on simple meshes. You can also always zremesh the merged mesh another time to get rid of any triangles that might appear around the neck for clean topology.
Whether you use dynamesh or topology editing at some point topology editing is something worth picking up. It will save you having to buy topogun or 3D coat.
Dynamesh would work as long as it connects them into one solid polygon island, instead of two or more floating ones.
Otherwise you can use this tool to smoothly connect different pieces together. It’s like zremesher, but uses it to bridge the gap.
http://docs.pixologic.com/user-guide/3d-modeling/sculpting/sculpting-brushes/curve-bridge/
That seems like a lot of steps. I don’t entirely understand why I would need to create so many copies and do all that stuff just to connect two tools together. I mean, I admit that I don’t really understand the way ZBrush works with most things, but in 3DS Max, I would just need to use the weld tool or bridge tool to do the same thing. The only problem is that it would then mess up my UVs and I would have to unwrap it and texture it again, which is what I want to avoid. So why does ZBrush require all of those steps?
I think would help to narrow down the process if we see the model. From start a head normally doesn’t need to be connected to the clothing. As you say it needs, it is difficult to predict what exactly are the needs of your project and then optimize the workflow.
so why does ZBrush require all of those steps?
It doesn’t. Select the curve bridge brush, select the first border (head), then select the second border that you want to connect it to. Done.
Granted this wont preserve the UVs; this is something you can’t really do in zbrush when you start changing the vertex order around by adding new faces. If you absolutely need to preserve UVs then you could pipe the result to an external program that can transfer UVs between different meshes by matching the position of the closest vertex (an example being the GATOR tool in Softimage).
If it’s just a case of not really wanting to spend the time re doing the UVs, then UV master might speed that up.
Okay, so here’s the finished head…
And here’s a quick sketch of the clothing I wanted to do. Basically, just some simple renaissance clothing.
But when I modeled the neck, I just did it to vaguely indicate that it isn’t a floating head. Now, it needs to actually come down and look like a proper neck, and somehow look like it goes under the clothes. I thought that maybe the best way to do this would be to use ZSpheres to make the body, attach the head to it, and then use my standard brush to pull the clothing shape out from the body, rather than having everything be a separate piece. I’m not sure if that makes sense, or if that’s the best way to do things.
I also don’t really know about the legs. I want there to be boots that you can see under the skirt, and I want to be able to model this character in a T/A pose, and then actually rig and pose this character later. So do I model legs under the skirt? Or do I leave them out, since you can’t see them anyways?
You only need to extend a bit the neck. If you want a quick work you could even simply extend a bit the geometry moving it. Currently it is a flat color and I don’t think any stretch would be visible.
If you more geometry and stretch is not enough it depends of the current status of your geometry. That model looks polypainted. That won’t be lost, I think. If you have subdivisions you could freeze the subdivisions, extend the model with zmodeler and unfreeze. If the polypaint has been affected you could keep a copy a re project the polypaint. Some extra modeling will be needed to smooth the new geometry.
If I’m modeling it I certainly would not connect the head to a full body. Leave the clothing disguise any seam. You could use the geometry of a body as reference for volume and proportions.
I’m a little concerned with keeping the head separate, simply because in the past, this has caused a few clipping problems when I went to animate the model. Basically, the bottom area of the neck would come up and show through the shirt. However, that was in 3DS Max, and I’m not animating this one, so I’m not sure if I’ll have the same problems in ZBrush?
No, you won’t have problems in Zbrush