This small readme file has only some basic info on the code structure. I didn't write an extensive documentation, and while I did spend a lot of time on this project, please bear in mind that it is a research project. As such, the code might be a little messy, and not always easily understandable. I'm sorry about that, but heck, you're getting it for free! However, if you have small questions, I'd be happy to answer them, you can find me on G+ (http://gplus.to/fxthomas), Twitter and LinkedIn (although I'm much more likely to answer on G+).
You can copy/paste/modify my code however you want, but I'd ask that you put my name in the credits, that's a small way to say "thank you!".
This is a small research project about understanding, implementing and testing a paper by Hamed Pirsiavash, Deva Ramanan and Charless C. Fowlkes called "Globally-Optimal Greedy Algorithms for Tracking a Variable Number of Objects". You can find my project report on the "Downloads" section of this GitHub repository, as well as on the "Report" directory of the project.
The data I used in this project came from two sources, so feel free to check them out :
- Part-Based Object Detector : http://www.cs.brown.edu/~pff/latent/
- ETH Moving Sequences Dataset : http://www.vision.ee.ethz.ch/~aess/dataset/
- Python 2
- NetworkX
- SciPy, NumPy and Matplotlib
- MATLAB for running the object detector (some results are included in the seq_ethms_results.mat file)
That's all folks! See my report for more, and hope you like it!