/lucas_kanade_tracker

Lucas Kanade Tracker (Part of ENPM673 - Perception for Autonomous Robots course)

Primary LanguageJupyter NotebookMIT LicenseMIT

Lucas Kanade Template and Inverse Compositional Tracker

Authors

Shantam Bajpai Arpit Aggarwal

Example of Lucas-Kanade Tracker on Car Video

Instructions for running the code

To run the code for Bolt Video, follow the following commands:

python Code/main_bolt.py 'images_path'

where, images_path is the path for bolt images. For example, running the python file on my local setup was:

python Code/main_bolt.py data/Bolt2/img

To run the code for Car Video, follow the following commands:

python Code/main_car4.py 'images_path'

where, images_path is the path for car images. For example, running the python file on my local setup was:

python Code/main_car4.py data/Car4/img

To run the code for Dragon Video, follow the following commands:

python Code/main_dragon.py 'images_path'

where, images_path is the path for dragon-baby images. For example, running the python file on my local setup was:

python Code/main_dragon.py data/DragonBaby/img/

Software Required

To run the .py files, use Python 3. Standard Python 3 libraries like OpenCV, Numpy, and matplotlib are used.