Skip to content

dvp-tran/TrackingHMM

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

61 Commits
 
 
 
 
 
 
 
 

Repository files navigation

TrackingHMM

Tracking example

Idea

Face tracking using HMM and particle filtering based on facial contour and skin color.

How to use

You can launch the project with the main.py script in the src folder.

Source

With the --source argument you can specify the source of the video :

  • Pre-loaded video sequence available ine the data folder, for example --source sequence1
  • Webcam live feed : --source webcam

Display

If you want to display the full particles and cercles, use the --fulldisplay argument.

Example

  • sequence 1 with full display : python3 main.py --source sequence1 --fulldisplay

  • live webcam with standard display : python3 main.py --source webcam

Tweaking

  • You can play with the particle_number and std variables from the main.py script to adjust the particule number and the standard deviation of the motion equation.
  • You can also play with the threshold variable in the observation.py script to play with the threshold of likelihood a particle must have to be considered as skin color.

Requirements

This scripts runs on Python 3 using the following libraries :

  • scikit-image
  • opencv
  • matplotlib
  • numpy
  • scipy

Contributors

Pascal Jauffret, Alexis Rosuel & Duc-Vinh Tran

About

Face tracking using HMM

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 92.8%
  • Python 7.2%