Researchers at Carnegie Mellon University's Human-Computer Interaction Institute (HCII) can determine when a standard smartwatch wearer is typing on a keyboard, washing dishes, petting a dog, pouring from a pitcher, or cutting with scissors.
After making a few changes to the watch's operating system, the researchers found they could use its accelerometer to recognize hand motions.
In addition, in some cases, the researchers could identify 25 different hand motions at about 95% accuracy using bio-acoustic sounds.
Said HCII researcher Chris Harrison, "A wide variety of apps could be made smarter and more context-sensitive if our devices knew the activity of our bodies and hands."
From Carnegie Mellon University
View Full Article
Abstracts Copyright © 2019 SmithBucklin, Washington, DC, USA