Progress Report - 6 Jul 2007
1. What were my goals for this week?
- Get ideas for clustering algorithms
- Do manual clustering of the motion to determine the potential effectiveness of affine fitting
- Look at how to pick a representative point
- Determine a good vs. bad fit of affine transforms
- (long term) Learn Illustrator
2. What did I accomplish this week?
- Implemented manual clustering to strive for a better fit, tried with 2, 3, and 4 divisions
- Wrote a quick method for the SpringNetwork program to spit out an edge connection matrix
- Organized BibTeX references for Mike and got familiar with JabRef
- Brainstormed on methods of picking representative point; did not try any of them yet
3. Why are numbers 1 and 2 different?
First and most unfortunately, I only had 3 days to work this week between a trip home to visit my parents and the fourth of July (neglected to transfer relevant files to an accessible folder, and the building was locked). Thus I didn't have time to start on the long term goal (learning Illustrator), and the implementations of my brainstorms didn't get done.
4. What are my goals for next week?
- Implement a couple different types of clustering, naive and hopefully something a little smarter
- Do a search through existing bibliographic tools and see what's out there and if we need to build our own
- Work on a good way to pick out "critical" points to represent motion without showing paths from every point
5. How does this fit in my bigger picture?
Hopefully this is a good basic place from which to build the final representative motion images!
6. What did I read this week?
- Har-Peled, Sariel, Clustering Motion.
- Ben-Hur, Asa and Horn, David and Siegelmann, Hava and Vapnik, Vladimir, Support Vector Clustering.