Kanade-Lucas-Tomasi Tracker

An implementation of the KTL tracker for the WASP course.

The details of algorithm is describe in the notebook

How to run

Install requirements (optionally into a virtual env):

pip install -r requirements.txt

Run python main.py -h for help.

usage: main.py [-h] [--custom] [source]

KTL Tracker

positional arguments:
  source      The source, defaults to webcame, else filepath to a video.

optional arguments:
  -h, --help  show this help message and exit
  --custom    use my custom KTL implementation.

To run using the webcam:

python main.py

To run on a video:

python main.py data/coke_zero.mp4

License

MIT