International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 12 Issue: 09 | Sep 2025
p-ISSN: 2395-0072
www.irjet.net
OMNI FOCUS OPS Amit Kumar
Meeniga Ranga Rahul
Musuku Chinna
Department of CSE Parul Institute of Engineering and Technology, Vadodara, Gujarat, India amitkumar33412@paruluniversity.ac.in
Department of CSE Parul Institute of Engineering and Technology, Vadodara, Gujarat, India 2203031050364@paruluniversity.ac.in
Department of CSE Parul Institute of Engineering and Technology, Vadodara, Gujarat, India 2203031050392@paruluniversity.ac.in
Gouru Rohith Chandu
PBSPR Vital
Department of CSE Parul Institute of Engineering and Technology, Vadodara, Gujarat, India 2203031050209@paruluniversity.ac.in
Department of CSE Parul Institute of Engineering and Technology, Vadodara, Gujarat, India 2203031050522@paruluniversity.ac.in
---------------------------------------------------------------------***-------------------------------------------------------------------A key feature of the proposed system is its integration Abstract— This paper presents Omni Focus OPS, a low-
cost, software-only eye-controlled mouse that uses a commodity webcam and MediaPipe Face Mesh to estimate iris/eye positions and map them to system cursor coordinates. The system performs continuous cursor movement, blink-based clicking and integrates a lightweight human-presence detector and anti-spoofing heuristics to reduce false activations. We describe the complete system design, mathematical mapping from image-space to screenspace, accuracy metrics, and an implementation that prioritizes real-time performance and minimal calibration.
of human detection and eye gesture recognition. The design supports both left and right clicks through controlled blinking patterns, while also differentiating between natural blinks and intentional eye gestures. This functionality is implemented using Python libraries and advanced detection algorithms, including Haar cascade classifiers and MediaPipe-based hu- man feature tracking. The approach is divided into three primary stages: Input, Processing, and Output. The input stage captures live video streams through the laptop camera. In the processing stage, deep learning techniques identify human facial landmarks and track eye movement with high precision. Finally, the output stage translates these movements into corresponding cursor actions, providing a smooth and responsive human–computer interaction experience.
Keywords: eye tracking, gaze cursor, MediaPipe Face Mesh, human detection, blink click, accessibility, Omni Focus OPS
1.INTRODUCTION The exponential growth of computing power has brought us closer to a future where technology can be accessed seamlessly without the need for traditional devices. This progress is shaping the evolution of device-less interaction, where human gestures replace conventional hardware for controlling digital systems. Such innovations not only improve convenience but also reduce e-waste, while providing transformative solutions for individuals with physical disabil- ities. Users with impaired motor functions but intact cog- nitive abilities can particularly benefit from these systems, gaining the ability to express their ideas more effectively.
Our contributions are: A software-only pipeline using MediaPipe Face Mesh for iris/eye landmark extraction and PyAutoGUI for cursor control. • A blink-based intentional click detector with simple anti-spoofing rules and a human-presence detector to avoid accidental activations. • Quantitative evaluation metrics (mean pixel error, click precision/recall, latency) and recommended experimen- tal protocol. •
Deep learning plays a central role in this advancement, enabling complex tasks such as image recognition, speech processing, and gesture analysis. In healthcare and assistive technologies, deep learning has already demonstrated its potential by supporting diagnosis and enhancing patient care. Building on this foundation, our work introduces an intelligent system that replaces conventional mouse input with real-time eye-tracking–based cursor control. The system leverages a standard laptop webcam to detect and map eye- ball movement into cursor actions, eliminating the need for external devices and thereby reducing cost and complexity.
© 2025, IRJET
|
Impact Factor value: 8.315
2. RELATED WORK Webcam and camera-based cursor systems have been a subject of research for many years. Prior work ranges from hardware-assisted systems (IR cameras, external trackers) to software-only solutions using OpenCV and classical image processing. Several recent studies demonstrate the viability of webcam-based eyeball cursor control and deeplearning approaches for enhanced robustness. Representative exam- ples include camera-based eyeball cursor systems that rely on Haar cascades and Hough
|
ISO 9001:2008 Certified Journal
|
Page 274