View Full Version : Real-Time Anthropometry Female Model
02 February 2004, 07:29 PM
I've posted this on other forums but I though I would show it off here and ask for any crits on areas to improve it. If anyone can tell me a good way to do the hair it would really help.
I'm building it for use in a real-time environment for validation of architectural standards and design. This is the Female 5th percentile model ( the smallest sized adult human, for use with testing reach and accessibility of controls and display.)
I have a target tri count of 3000 polies. So far the count is 2080 triangles, modeled it in Max 5.
02 February 2004, 10:09 PM
Looks like you've used the joan of arc tutorial reference material as the base for your modelling. That particular reference is a little disproportionate to the normal human figure, particularly the size of the head.
If you're going for something realistic you'll need some different reference material.
The area between the breasts is a little too well defined considering she's wearing a t-shirt... IMO they're also a little too close together, but that may be the fault of the reference material. Try making the curve in the small of her back a little less apparent as she might be hard to rig. You can always accentuate the curve later when she's all rigged up (If your going to pursue this model that far).
The crit may not be what your after, since your only using her as a reference, but what the hell... no one else has said anything so far.
02 February 2004, 08:35 AM
Actually I've heard that about this resmbling the Joan of Arc reference before. The only explanation I have for that is that I've actually used that referece sheet many time in the past as I liked the psudo-anime look it had. I guess I've just used it so often I could probably block the pose out in my sleep. But I do see it also, even though I haven't actually used any refrences except the data I have. The head may need some reworking, because it is misproportioned, and I'll play around with the breast alittle more (even though I tend to get in trouble for doing that kind of thing.)
The model will not be rigged, there will be 4 static poses for it. Standing, standing reach, sitting and sitting reach.
For this model though I've been using anthropometrics data. Basically a bunch of measurements taken from a large study group. Then the measurements are grouped and averaged together. The data is normally broken up into Male and female groups giving the 5th percentile, average, and 95th percentile sizes. Most people are in the average category. But when testing design and ergonomics the two extremes are used for validation. Normally the Male data is used for the 95th percentile due to the fact that human males are genetically disposed to being larger than females (at least in most western cultures, the data does vary from ethnic group to ethnic group.) This is the same reason that female are used in the 5th percentile size. As far the proportions go I've altered many of them to make the model look less box like or misshapen.
Here is a look at the data and how it fit's to the model.
I first created the matrix of plains before I begain modeling. I then blocked out the body and positioned the limbs (again I've used the joan tut so often that the referance pose seems most natural.) I have been using Ben's (Poopinmymouth) video tutorials to aid in the mesh flow though.
01 January 2006, 01:00 PM
This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.