Precision typing on a smartwatch with finger gestures (Kurzweil Accelerating Intelligence)

Precision typing on a smartwatch with finger gestures If you wear a smartwatch, you know how limiting it is to type it on or otherwise operate it. Now European researchers have developed an input method that uses a depth camera (similar to the Kinect game controller) to track fingertip touch and location on the back of the hand or in mid-air, allowing for precision control. The researchers have created a prototype called "WatchSense," worn on the user's arm. It would also work with smartphones, smart TVs, and virtual-reality or augmented reality devices, explains Srinath Sridhar, a researcher in the Graphics, Vision and Video group at the Max Planck Institute for Informatics. KurzweilAI has covered a variety of attempts to use depth cameras for controlling devices, but developers have been plagued with the lack of precise control with current camera devices and software. The new software, based on machine learning, recognizes the exact positions of the thumb and index finger in the 3D image from the depth sensor, says Sridhar, identifying specific fingers and dealing with the unevenness of the back of the hand and the fact that fingers can occlude each other when they are moved.