Progress Report - 22 Jun 2007

1. What were my goals for this week?

2. What did I accomplish this week?

3. Why are numbers 1 and 2 different?

Unfortunately didn't get much clustering work done at all - spent most of the time working on how to get a meaningful arrow out of the information we have. This took most of the week, seeing as there are a bunch of different methods for doing so. Also had to do some psychological case testing, consisting mostly of drawing out various types of arrows and figuring out what they would imply about the motion, in an attempt to try and get to the point where I would be communicating the appropriate things to the viewer.

This took most of the week, as the papers were pretty mathy, and not only did I want to try and understand how the decompositions worked thoroughly enough to be able to implement something, but also try and understand the A = x0\x1 operation thoroughly enough to implement it in C++, which I looked at for about half a day and then gave up on as a waste of time for now. (Also, spent Friday morning helping Greg with the images for his paper.)

Ran into a wall with Alexa's implementation - log(A) ended up with complex entries - until I figured out that there were specific logarithm and exponential implementations for matrices. Once I started using logm(A) and expm(A), things went a lot better. D'oh.

exponential interpretation of first frame

4. What are my goals for next week?

5. How does this fit in my bigger picture?

Get the Mitchell Lab stuff done so that I can concentrate completely on research!

6. What did I read this week?