Eye Interaction

Inspiration

Imagine waking up and realising you are not in your bed, but in a hospital. You try to move your body, but nothing responds. You try to speak, but no words come out. You are fully conscious, but completely trapped.

This is not science fiction this is the reality for people who suffer from brainstem stroke. They can see, think, and understand, but cannot control their bodies except for their eyes.

This reality inspired us to build a solution where the only remaining ability eye movement becomes a way to regain control and independence.

What it does

Eye Interaction is an AI-powered system that allows users to control devices using only their eye movements.

We developed:

An eye-controlled wheelchair for mobility A smart home system to control appliances A flexible platform that can be integrated into multiple devices

Our goal is to turn eye movement into a universal control interface.

How we built it

We used computer vision and AI techniques to detect and track eye movement using standard cameras.

The system processes gaze direction in real time and converts it into control commands. These commands are transmitted to devices such as wheelchairs or home appliances using embedded systems and wireless communication.

We focused on building a low-cost, scalable solution that works without expensive specialised hardware.

Challenges we ran into

Achieving accurate eye tracking in different lighting conditions Handling users with glasses or partial occlusion Ensuring real-time response without delay Integrating software with hardware systems like wheelchairs Making the system reliable for real-world use

Accomplishments that we're proud of

Successfully built a working eye-controlled wheelchair Developed a home automation system controlled by eye movement Received early support from UNICEF Presented the project internationally, including Startup Istanbul and academic platforms Generated revenue through related research and technical projects Impacted real users by improving independence

What we learned

Technology must be built for real users, not just for demonstration Simplicity and affordability are key for accessibility Real-world testing is critical for reliability Building something meaningful requires persistence and continuous iteration

What's next for Eye Interaction

Improve system accuracy and performance Conduct pilot deployments with real users and healthcare partners Expand into smart home and multi-device integration Explore IP protection strategies Scale through partnerships and licensing with manufacturers Expand into global markets including the UK, Europe, and the USA

Built With

• amazon-web-services
• c
• c++
• computervision
• machine-learning
• microcontroller
• python
• robotics