Your smartwatch’s incredibly tiny touchscreen isn’t necessarily the easiest way to navigate its interface. So to make hand gestures more reliable and more robust, a team of researchers has created a strap that can see inside the wearer’s arm and track the movements of the muscles instead.
The various muscles in your arm that make your hand and fingers move contract and expand in countless different ways in the process. So researchers at the Human-Computer Interaction group at Carnegie Mellon University, led by Chris Harrison and Yang Zhang, have created a special smartwatch strap that can visualize how those muscles move, and use those unique images and patterns to determine what hand or finger gesture is being performed.
The Tomo, as the prototype is called, features a series of electrodes around the inside of the smartwatch’s strap that each send a small electrical signal through the wearer’s arm, and then measure its strength coming out the other side. With enough measurements gathered, an image of the inside of the arm can be generated and analyzed.
It’s a technique known as Electrical Impedance Tomography—similar to PET and CT scans—but much smaller and far cheaper than the medical scanners used at hospitals.
The images generated by the Tomo strap are low-resolution, but still provide enough detail for software to distinguish between a wide variety of hand and finger gestures being performed. And because the analysis is all being done internally, the system should work just as well in the dark, or even when the wearer has gloves on.
It’s not perfect yet, though. Differences in how the strap is worn from day to day can throw off the software’s ability to accurate detect gestures as defined on a previous day. And since everyone would be performing them differently, with varying physiologies between users, the Tomo system would have be taught to recognize the gestures first, which would make setting up a new smartwatch and even more agonizing procedure.