Even today, keyboard remains one of the main ways of Communicate with a computer. Over time, theHuman-machine interaction Significantly evolved. However, this habit has not changed. Moreover, there is a recent invention that reinforces this trend. In fact, researchers recently found a file to be more efficient When interacting with machines. It’s about Hand gesture recognition.
Currently, this new technology is still in its early stages. development stage. The researchers who designed his name Sample “similar to a pattern”. To accomplish its function, the program uses an extension laptop web camera Normal. Recognizes hands near the keyboard and performsHand positioning process.
Typealike is a file revolutionary technology which is based on a artificial intelligence. According to the designers, it can be exploited in the field of Virtual Reality. In this sector, the program will have the potential to Eliminates the need for manual controls.
Typealike Origins
according to Nalin Chipper, the group of researchers first had an idea to look for New webcam uses. Note that Nalin has a master’s degree from the University of Waterloo’s Chereton School of Computer Science.
“The webcam is aimed at your face, but most of the interaction on the computer is around your hands. So we wondered what we could do if the webcam could pick up on hand gestures.”
Nalin Chipper
From this idea, Team A invented mechanical attachment able to Redirect the webcam towards the hands. Subsequently, the researchers developed a A program that can distinguish many hand gestures. It should have been able to recognize an accurate gesture under several circumstances, for different users. In order for it to be effective enough, they had to train it.
Artificial intelligence program
Then the group of researchers machine learning techniques at the time of Pattern-like training. In other words, the designers developed a file artificial intelligence specific. Fabrice Matulich, Principal Investigator at Preferred Networks Inc. and a former postdoctoral fellow at Waterllo, further elaborated on this topic.
Since it’s a file neural network, they had to train the algorithm to Detection of different positions of the hand. In addition, the shape and sizes of the hands can affect the user’s gestures. As a result, they collected a lot of data From several people in varying lighting conditions. This machine drive was made possible by Dozens of volunteers.
