Eye Gaze let user experience real-time Video Stabilization and Gaze Tracking
Eye Gaze is the application developed for my master's thesis. It is a simple demonstration of the implemented features.
In particular, the app lets you to stabilize (for small displacements) the camera shakes and track the gaze direction after the detection of the user's facial features.
The developed system exploits the OpenCV library.
VIDEOSTABILIZATION:
There are two possible approaches to video stabilize the camera, but not every device support both adequately.
TRACKING:
In order for the system to work, the user must focus the face. Furthermore, wearing glasses or the presence of disturbing elements modifies the application's right performance.
For the system's optimization, you should put yourself in adequately illuminated places. As a matter of fact, too much illumination or an inadequate one could modify the system's performance.
The facial feature which is considered the most influencing one on the system's correct performance is the eye's extremity, especially its detection; this is the reason why the user is free to choose the approach which he considers to be the best one.
**** INFO ****
For any info, please contact me via e-mail
Download