Researchers have shown that smartwatches can track your daily activities to the minute

Future privacy nightmares could be brought on by wearables. It has been demonstrated that a straightforward software upgrade can enable businesses to determine precisely what activities a wearer is engaging in.

Exercises including walking, running, biking, swimming, and sleeping can be evaluated by smartwatches and fitness trackers. However, researchers at the Human-Computer Interaction Institute (HCII) at Carnegie Mellon University have shown that such gadgets can be significantly more context sensitive.

Over the course of two weeks, they enlisted 50 participants to wear typical smartwatches, and then observed them as they went about their everyday routines. The wearer’s activity was inferred using data from the accelerometer and, in some circumstances, bio-acoustic sounds collected by a specialised programme on the watch. The next step was to ask the subjects to characterise the current hand activity.

With the help of this method, 25 separate activities may be identified with greater than 95% accuracy. They consist of using a keyboard, doing the dishes, twisting jars, caressing dogs, brushing teeth, pouring from a pitcher, and other activities. A novel dataset was presented by the identification of about 80 hand activities in total.

Numerous applications could be made of this knowledge. For instance, smartwatches might prevent interruptions when someone is slicing vegetables or operating heavy machinery, similar to how smartphones can block communications when a user is driving.

These details could also be used to encourage people to stop harmful behaviours like smoking or ensure that they frequently practise healthy habits like brushing their teeth. There are countless options.

The paper says that hand-sensing “may also be employed by apps that provide feedback to users who are learning a new skill, such as playing a musical instrument, or engaging in physical therapy.”

Of course, the need that the user wear the smartwatch on their dominant arm is a restriction. Such gadgets are typically worn on the arm of the wearer who does not actively participate in daily activities.

According to Chris Harrison, assistant professor in Carnegie’s Human-Computer Interaction Institute (HCII), “we view smartwatches as a unique beachhead on the body for recording rich, everyday behaviours.”

If our devices were aware of the movement of our bodies and hands, a wide range of apps could be made smarter and more context-sensitive.

However, this does raise some privacy issues. If watches become too intelligent, they will be able to track your daily activities. Most people wouldn’t be very pleased about that, though.

This week, the research will be presented in Scotland at “CHI 2019,” a conference on human factors in computing systems organised by the Association for Computing Machinery. Take a look at the video below; it’s entertaining.

Leave a Reply