Markov chains. Markov fields. Wherever I go, they follow me. The reason I took the course was a rather face-my-fear sort of reasoning. Scared the shit out of me, to be honest. But there were good(!!) things. The last assignment we had dealt with MRF applications in vision. Texture synthesis and parameter estimation to be precise. It was a b***h of an assignment, but I had to read up a lot for that. Which is never a bad thing to do. Anyway, I'll be putting up stuff relating to this course from now on. First up is a literature review that I did on Sequential Monte Carlo methods for tracking; i.e. particle filtering and it's different flavors. It was done on five paper, as listed in the reference. The review got 90% grades, so I assume it's not too bad a paper to put up for other peoples help.
Literature Review |
Sequential Monte-Carlo methods on tracking single and multiple objects. Graded 90 out of 100, with the comment that a little more detail on partitioned sampling would be nice. |
|
Term Project |
An implementation of Xu and Prince's algorithm for active contours or snake formation using gradient vector fields. The writeup contains the algorithm and results. I've put the tex file here as well, you might want to use it as a template. Graded 100% (not that it means anything at all). |
|
| |
|
|
|
