Eye-gesture control of computer systems via artificial intelligence

Area of Science:

• Computer Science
• Human-Computer Interaction
• Artificial Intelligence

Background:

• Artificial Intelligence (AI) presents significant opportunities for advancing human-computer interaction.
• Eye-gesture recognition offers intuitive control and improved accessibility for individuals with physical impairments.

Purpose of the Study:

• To develop and evaluate an AI-driven eye-gesture recognition system.
• To translate eye movements into actionable commands for enhanced user control.

Main Methods:

• Developed an AI system utilizing OpenCV, MediaPipe, and PyAutoGUI.
• Trained the system on 20,000 gestures from 100 diverse volunteers.
• Tested system performance under varied lighting and eyewear conditions.

Main Results:

• Achieved 99.63% accuracy in general gesture recognition.
• Demonstrated robustness with 98.9% accuracy under reflective eyewear.
• Confirmed system adaptability and generalizability across diverse scenarios.

Conclusions:

• The developed system significantly advances AI-driven interaction through enhanced accessibility.
• Potential applications include critical fields such as military and rescue operations.
• Future research will focus on validation with public datasets to broaden impact.