MakeHuman

From Wikipedia, the free encyclopedia

MakeHuman

MakeHuman 0.9.1 RC1 Promotional Art "Occhi" (Eyes).
Developed by The MakeHuman team.
Latest release 0.9.1 RC1 (Release Candidate 1) / December 2, 2007
OS Windows, Linux, Mac OS X
Genre 3D computer graphics
License GPL, MIT
Website http://www.makehuman.org/

MakeHuman is a software application that generates 3D humanoids; similar to Poser or DAZ Studio. It is written in C++ and Aqsis is necessary to produce a render. The MakeHuman team work towards correctness both in programming (using common file formats) and anatomy. MakeHuman makes extensive use of university research in accurately modelling the human form.

Contents

[edit] Features

  • Poses engine: a system to simulate muscular movement and limb rotation
  • Facial expressions
  • About 3000 morphings
  • System of mixing based on Sheldon's anthropometric technics (endomorph, mesomorph, ectomorph)
  • Skin rendering with Renderman compliant engines.
  • Autorigging: auto-adapting to deformations of morphings
  • Morphing alchemy: system of mixing different characters
  • Wavefront Obj export.
  • Collada export.

[edit] History

The MakeHuman ancestor is Makehead, a Blender script written in Python by Manuel Bastioni. It was based on the Blender vertex keys and modelled only the character's head.

MakeHuman started as a post from Manuel Bastioni on the kino3d forums in the year 2000. Four months after this post Manuel Bastioni, Filippo Di Natale and Mario Latronico wrote the first alpha of MakeHuman, it was released under GPL and it was able to generate a whole humanoid. Enrico Valenza modelled the base mesh version 1. This mesh is used to generate the humanoid using parametric morphing targets. At this time the project was hosted at the Blender site that offers an infrastructure for projects (mailing lists, forum, cvs, bug tracker) related to Blender.

For version 1.6 (8 January 2004[1]), the old and complex GUI was totally rewritten and the site moved to dedalo3d. The version 1.8 (5 April 2004[2]) was written with the help of new members, Craig Smith as a modeller, Oliver Saraja as a coder, Michael Schardt wrote the import-export of vertex groups and Andrew Kator the system of bones.

Paul Ekman, a psychology teacher donated 20 years of research known as FACS (Facial Action Coding System) to the MakeHuman team. This allowed the facial expression features of MakeHuman.

The version 2.0 of the Python script (that never exited the alpha stage) had a completely new base mesh (11.000 vertex) modelled by Kaushik Pal, because the old one (7.000 vertex) lacked some edge loops for the facial and muscolar animation. The new base model was presented to famous modellers like Steven Stahlberg and Tamás Varga to have critics.

MakeHuman 0.8a (released on 3 November 2005) was at this point completely rewritten in C by Paolo Colombo, this new MakeHuman had a new interface, was multiplatform and independent from Blender.

Screenshot of MakeHuman 0.9.1 RC1 showing 'tetrawidget'.
Screenshot of MakeHuman 0.9.1 RC1 showing 'tetrawidget'.

MakeHuman 0.9 was released on 1 November 2006. Andreas Volz, a student of applied computer science at the university of Fulda in Germany developed a library in C++, called Animorph, for his bachelor thesis called "MeasureHuman". After he received his degree he published his library under LGPL and this library was adopted by MakeHuman. The MakeHuman team and the Ninibe Labs wrote a new GUI (MHGUI) in C++ using OpenGL and object oriented design principles.

On the 2nd of December, 2007 the MakHuman project turned another corner with a new website, a new logo, and also a new release - 0.9.1 RC1. The most notable difference between this and the previous release is the addition of the 'tetrawidget' - "an interface with just 4 buttons with which 90% of the characteristics of the human body can be modeled".[3]

[edit] See also

[edit] Notes

  1. ^ Blender Projects: makehuman-1.6
  2. ^ Blender Projects: makehuman-1.8-download
  3. ^ MakeHuman website. [1]. Visited June 10, 2008.

[edit] External links

Languages