/ultraking

OpenCV + Python implementation of the paper "Incremental Learning for Robust Visual Tracking"

Primary LanguagePython

Ultraking

OpenCV + Python implementation of the paper Incremental Learning for Robust Visual Tracking

Dependencies

  • a webcam 🎥
  • Python 2.7
  • OpenCV 2.4.x
  • python-opencv

As described further, you can either use the virtual machine provided by Vagrant (recommended) or install manually all the dependencies.

Play around

Launch the pre-configured virtual machine

  1. Install Virtualbox and the extension pack (needed for the webcam): https://www.virtualbox.org/wiki/Downloads

  2. Install Vagrant: https://www.vagrantup.com/downloads.html

  3. Start the pre-configured graphical VM with:

    vagrant up

  4. Attach your webcam to the VM through the Virtualbox menu Devices > Webcams.

  5. Open a terminal in the VM and go to the shared folder:

    cd /vagrant

Run Ultraking

python ultraking.py

Usage

  1. Select the part of the image you want to track with the mouse (left click and drag)
  2. Press Enter to start the tracking
  3. Have fun
  4. Quit by pressing a button (all except Enter)

Resources

Incremental Learning for Robust Visual Tracking, http://www.cs.toronto.edu/~dross/ivt/RossLimLinYang_ijcv.pdf