1. Subhadip Ghosh, Leslie Watkins, Sajika Gallege, Sean Andrist
2. Tools: C++, MEL, Maya API, possibly some Python for writing converters
3. Plan: At the end of this infrastructure part of the project we are hoping to have a Maya plug-in that will meet all of the requirements listed. For now, Leslie and Subhadip are concentrating on learning how to read in skeletal motion capture data, while Sean and Sajika worry about the marker position data. Along with that, Sean is concentrating on figuring out the Maya side of things, while Subhadip, Leslie and Sajika begin coding the C++ data structures and Python converters to get the motion data in an internally consistent format.
4. Signs of life: Unfortunately we don’t have anything graphical to show yet. The group is working on identifying appropriate data structures, reading in the data, finding nice converters so we don’t have to worry about so many file formats, and getting the Maya API for C++ set up in Visual Studio. We also set up our mercurial repository today, which the whole group found to be pretty exciting. By mid next week we hope to start dividing up the work for the motion splicing, interpolation, etc.
{ 1 comment }
The Maya plugin might be ambitous: not that its more work, but that having to build and test and debug things with Maya around might add a layer of complexity. (does MEL even have a debugger?) You might try to have pieces that can work on their own for testing but then can be plugged in.
There is a “modern” version of Python for Windows in /p/graphics/Tools/python27 – (with lots of packages installed). You may want to copy this tree to your local disk (although, if you want this, let me give you the ZIP file since it takes AFS a long time to deal with the zillion tiny files).