Providing universal topology for physically accurate deformations in existing realtime game engine workflows.
The primary purpose of this project is to create a feature-rich human 3D model which can be adapted to any relatively-realistic human form (regardless of gender, ethnicity, age, etc). By adapting the shape of the model without altering its topology, all features like physically accurate rigging, edge flow, and deformations are preserved (and transferrable). In addition to physical accuracy, universality, and scalability, it's also a priority for this model to by fully compatible with existing realtime rendering and game engine standards.
The first region of interest to be developed in this project will be the human face. Stay tuned...
A concerted effort was made to represent distributed human diversity amongst the selected reference images in order to best inform a universal human topology. Unfortunately, the images freely available in the public domain (or equivalent) disproportionaly represent certain human populations over others. There was also understandably a bias towards selecting images which would be informative to human topology, such as visibility of pores/creases/wrinkles, strong or unique muscle activation, an unobstructed view of the skin, etc. As a result, despite all efforts to construct a universally inclusive basemesh, this model's topology may represent a skewed human demographic.
If you identify a lapse in universal topology, please open a public issue or email me privately via [email protected] to discuss incorporating the underrepresented features into this project.
This work is made freely available under the CC0 1.0 Universal public domain dedication, the text of which should be found in COPYING.txt in the root directory of this project.
No additional/conflicting permissions were present in the source repository at the time of release.
Repository: https://gitlab.com/recursivenomad/biomechanical-basemesh/
Contact: [email protected]