In this week we setup the different metrics used and get used to the dataset we used in the following weeks.

The main goal of this week is to estimate the background in a video sequence using different statistical models and compute the metrics to compare them.

Detection using Deep learning networks and comparing tracking methods.

This week goal was to evaluate the contribution of Optical Flow in video stabilization and tracking. comparing different method to obtain the motion field.

In this final week the goal was to solve the multi camera multi tracking problem. We implement three methods and report the metrics for the best of them. You can also find a report of this week under the folder report.

The code is structured in the following way:

• src folder with the sources code. Inside this folder, you will find:

  • weeks package, with one package created for each week.
  • utils package, with different modules of useful functions (Work in progress).
  • metrics module, with different metrics to measure the performance of our experimentss (Work in progress).

• data folder contains the data provided with the codes (Work in progress).

• output folder contains the output of each week. Each subfolder corresponds to each week.

To run each task you need to uncomment the line from the src/main.py file.

To run the files, you need to install the dependencies listed in the requirements.txt file:

$ pip install -r requirements.txt

Or you could create a virtual environment and install them on it:

$ mkvirtualenv -p python2.7 m6
(m6) $ pip install -r requirements.txt