- Projects 4
- Followers 3
Computer Syndrome
Computer Syndrome is a common challenge (Disease in advanced cases) for most people who spend their time on electronic devices. What if we offered a solution to solve this using Artificial Intelligence!!! This would mean more healthy people using electronic devices at optimum level. ...learn more
Project status: Under Development
Groups
Student Developers for AI,
DeepLearning,
Movidius™ Neural Compute Group
Intel Technologies
AI DevCloud / Xeon,
Intel Opt ML/DL Framework,
Movidius NCS,
Intel CPU
Overview / Usage
The common syndrome cases are computer vision syndrome which affects most daily computer and smartphone users. This project aims to use Deep Learning to create an application that observes the facial expressions of the person in front of the computer and detects when a person is either focusing for long hours without a break or is distracted for a long time. To reduce fatigue and other issues characteristics from the Computer Syndrome, the system will display an alert reminding the user to take a break or automatically take control of the application (users can allow using preference settings). Equally, if the user has been distracted for long periods of time the system will also emit an alarm and propose a solution. Customary users tend to be distracted by social media and increase their exposure to issues related to Computer Syndrome, as much as when they are focused on their work.
Methodology / Approach
To achieve a solution for computer syndrome, we aim to develop an AI application that can learn user behaviour and guide the user on safe and optimal techniques of using the device.
How to:
Implement an AI application using pytorch deep learning framework to learn user behaviour from the computer sensors (camera, eye focus and applications running or on focus).
We will be using train the model using the AI DevCloud, which will be deployed in the Intel Movidius Neural Compute Stick.
The application will keep a record of user behaviour and use this data to progressively improve its understanding of the behavioural patterns associated to each user.
Collaborators