Dear Anatomist,

Many demands have been received to create a female model.
It is obviously on the to-do list but it raises specific technical issues.
This post gathers the information about this aspect of the project.
How can a female model be created?
First, the male model is a modified version of 'BodyParts3D', created under CC-BY-SA license around 2012 by a Japanese team.
The technique used to produce this model is a mystery: the original files seem to be generated from a segmentation of a .dicom (scan/IRM), but the body is in perfect position (no flesh squashing due to the horizontal position). The question has been asked to the author and received no answer for the moment.
The visual quality of the atlas comes from this unknown technique: the overall shape is very natural and the fibers of each muscle are visible in the 3D mesh (the downside is that they are therefore heavier, and less texture-compatible).
In order to create an anatomically accurate female model, we need:
1) A .dicom reference:
It may already exist or have to be created.
Getting a detailed .dicom of the whole body is somewhat unusual but the main obstacle is, in my opinion, to find out/decide which is the best technique to produce a 3D model in anatomical position (standing) from a scan/IRM done lying down.
2) Create or adapt the meshes to fit this reference:
My proposition would be to transform the male model using the 'Lattice' modifier.
All the objects, the names and the relationships of the common anatomical structures would be already available and could be influenced together.
Much manual edition of the meshes and curves would then be needed to reach the final result.
The final result could then be:
-either another .blend file
-or the same .blend file, with the same objects and meshes, but in other 'Shapekeys'
This choice is a technical decision to take, based on the file size and memory (RAM) needed to navigate in the model (in Blender and in the mobile app).
All this represent an important amount of work and more workers/resources/time would be needed.
A solution would be to create a 3D model of the female anatomical structures and make a separate atlas of it. All the functions would be available.
It could then be integrated later on, if/when the conditions to produce the whole open source female model are gathered.
Back to the CONTRIBUTORS'HOMEPAGE