OCULAR CURSOR NAVIGATION USING PYTHON

Abstract

This project investigates the creation of a system that controls a cursor using the motions of the
human eyeball. It utilizes Python and takes use of the features provided by the OpenCV, MediaPipe, and
PyAutoGUI libraries. The main objective is to develop a hands-free, intuitive approach for human-computer
interaction that may greatly improve accessibility for those with physical limitations. The system utilizes
OpenCV for fast image processing and MediaPipe for accurate eye-tracking and facial landmark identification,
enabling real-time video recording via a webcam. Through the analysis of visual inputs, the system identifies
eye movements and converts them into matching cursor actions using PyAutoGUI. This process efficiently links
the direction of gaze to specific screen coordinates.
The implementation prioritizes three crucial elements: attaining exceptional precision in eye-tracking,
guaranteeing promptness in cursor motions, and maintaining a user-friendly interface for effortless use. OpenCV
manages the essential functions of video capture and preprocessing, which include preparing the data for further
analysis. MediaPipe, renowned for its sophisticated expertise in facial landmark identification, offers the
essential accuracy in monitoring eye movements. PyAutoGUI is used to translate these motions into cursor
actions, facilitating smooth control over the computer interface.
Preliminary testing and assessments show encouraging outcomes, with the system exhibiting accurate, seamless,
and instinctive engagement. The technique of eye tracking allows users to efficiently manipulate the cursor by
using their eye movements, demonstrating the promising possibilities of this technology in practical scenarios.
Subsequent efforts will be directed on enhancing the precision of gaze detection, minimizing delays, and
guaranteeing the system's reliable performance under various lighting conditions and with diverse user
demographics.
This project not only tackles significant accessibility obstacles but also demonstrates the actual use of modern
computer vision and automation technology. This system strives to enhance the entire user experience by
integrating robust libraries and prioritizing user-centric design. It specifically targets users with physical
limitations, providing them with a dependable and efficient solution. By doing so, it promotes inclusivity in
technology.