Research Update

Time for an update. I've learned from experience that I work best after careful research and planning, so I'm perusing a number of papers and websites related to my topic. To make things easier, I have divided my research topics into categories:

1) Human physiology (phenotypes, averages, muscle locations).

This research topic hasn't been my top priority because I have a large number of artistic references already, mostly books (Eliot Goldfinger, Vlippu). However, given that these references are stylized, I need to look up more realistic data - preferably medical.

2) Picking out cloned agents in crowds.

C. O'Sullivan spoke about the Metropolis project here at Penn, and her studies implied that on the face when differentiating figures in crowds. However, the purpose of my project is to define people by body phenotype when facial detail is lost. I feel this information gives a psychological starting point to my research.

3) Current crowd simulation technology.

This includes creating, texturing, and animating crowds efficiently. I need to know how my models will fit into existing crowd simulation software. Much of the existing information leans towards animating (behavioral, annotated environment, motivation etc.) and rendering crowds, not modeling. Currently, I am reading Crowdbrush, but I'm looking for more sources. I also have also found a paper about rendering clothing using level-of-detail which will probably be of interest when combining props and figure models (again, O'Sullivan).

4) Writing Maya plug-ins.

While I'm incredibly comfortable using the software artistically, I need to figure out how to integrate my project with Maya. I've been able to locate some sites with advice, but this is mostly related to learning the Maya API. I believe I will wrap C++, but I am also looking up resources on MEL, just in case. Specifically, I need more resources on writing and reading OBJ and BVH files, the general file formats which I want to integrate into the plug-in.

5) Uses and comparisons of crowd simulation technology.

I have a lot of papers tagged which relate to the validity of crowds, especially in cultural heritage sites. While not as directly related to my modeling topic, I feel I should research and tailor my plug-in for potential uses (like people who aren't familiar with coding). In addition, a number of these papers also cover realism of crowd simulation which relates psychologically to my work.

I have a wide breadth of abstracts now, so it's time to begin filtering them and reading more in depth. Once I have a better understanding of current technology and algorithms, I will work on understanding the Maya plug-in API. Once that is taken care of, I will be able to diagram my software.