What it does
O-Dot is a hands-free human-computer interaction system that lets users control a screen with their eyes and a smart ring. Designed for accessibility, multitasking, and lazy computing alike, offering a natural and minimal-effort interface.
Your inspiration
Modern computing relies on tools like keyboards and mice, but that doesn't work for everyone or every situation. Whether it's someone with mobility limitations, someone driving, or someone simply lying back watching Netflix, traditional input methods can feel restrictive. I wanted to break that wall. The idea came from noticing how often I needed to interact with screens but couldn’t (or didn’t want to) use a mouse or touchscreen. It made me wonder: why not just look at the thing, and tap? This evolved into a hybrid hardware and software system that reimagines how we engage with digital interfaces.
How it works
O-Dot consists of two parts: 1. Software that runs on your device and uses a webcam to track your eye movement. A lightweight machine learning model is trained on your gaze during a short calibration. Once trained, it can accurately detect where you're looking on the screen. 2. Smart Ring worn on your finger, embedded with three buttons. These act as mouse inputs: click, scroll, or drag. Instead of moving a mouse, users simply look at a screen element and press a ring button to trigger the action. This system enables full cursor interaction, without hand movement or physical navigation, blending accessibility with innovation.
Design process
Initial tests used open-source gaze tracking libraries paired with basic Python models for prediction. Over time, I improved calibration with personalized gaze mapping and temporal smoothing for stable cursor control. The ring was prototyped using an ESP32 microcontroller and tactile buttons, with Bluetooth Low Energy for communication. Hardware miniaturization was a major design constraint, along with achieving low-latency gaze prediction using just a webcam. I conducted iterative testing in real-world use cases, long study sessions, watching videos, and even designing code UIs, to ensure smooth interaction. Future iterations may include haptic feedback or gesture-based control via IMU sensors.
How it is different
Most gaze control systems require expensive eye trackers or rely on single-point interactions. O-Dot uses commodity webcams and adds multi-modal input via the smart ring. It doesn’t just click where you look, it lets you scroll, drag, or even customize actions. It’s not just a control tool, it’s a new interface paradigm. Light, intuitive, and modular.
Future plans
Next steps include improving calibration speed, adding predictive intent (to reduce false clicks), and refining ring ergonomics. I aim to open-source a version for accessibility communities and explore partnerships to bring it to users with motor disabilities, developers, gamers, and hands free productivity tools.
Share this page on