Microsoft is Developing a Gaze Tracker That Will Work on Every Device
Tracking where a person is looking is very useful for people living with motor neuron diseases and disorders.
Tracking where a person is looking is very useful for people living with motor neuron diseases and disorders.
of a person follow where you lookIt can be very useful for people living with motor neuron diseases and disorders. Because such a technology has the potential to help those people control their environment and communicate with others.
But predicting where a person is looking involves factors such as head posture, head position, eye rotation, background noise, and other factors such as glasses, face coverings, and ancillary medical equipment. due to variables It's not something that can be done easily. However, there are some commercially available devices that can predict where a person is looking.
Microsoft is developing a new look tracker:

The devices are available in the market, but the cost of these devices is up to thousands of dollars can reach. This is the situation in daily life, Researchers at Microsoft pushed him into a new struggle. Researchers, Works with any camera, hardware independent And extremely sensitive began to develop a gaze following.
A new report published earlier this month in the article The team shared the details of their work on the system. According to the shared details, researchers to work on any device The system they developed is in the GazeCapture dataset. 1.8073 centimeters margin of error had. Moreover, this result is not yet while not calibrated in any way was taken.

Microsoft has conducted gaze tracking research before. of the company published in 2014In the study, researchers tried to achieve results with cameras and depth sensors along with multiple infrared lights around a screen for eye tracking. The company's Windows 10 operating system, 'Eye ControlIt was the first operating system to offer the ' feature.
Researchers at Microsoft, in their latest article, published trigonometric models that focus only on the pupil and iris. detects the movement of other muscles They stated that it would not be enough. At this point, researchers using deep learning They said they could improve accuracy.
Every business is unique. Let’s explore how we can tailor these ideas to your own digital project.