Vision Based Communicator

This work provides a vision-based human-computer interface. The interface senses and interprets eye-blinks as controls along with gaze detection methods. We used image processing methods like haar-like features for automatic face detection, followed by eye tracking and blink detection based on landmarks. The gaze detection mechanism helps in easy controlling of the mouse pointer which can help people with disabilities to communicate effectively. A heatmap is also generated so that in future developments, this data can be used to design a more efficient layout for the controls. Given the necessity to move towards online solutions, such data can cater to the needs of a variety of sectors.