A pair of AI builders turned a $10 Webcam right into a motion-tracking system. It’s like a DIY Kinect with a Google mind. And better of all they named it Skeletron.
The challenge, a collaboration between Or Fleischer and Dror Ayalon, was developed utilizing Tensorflow, an open supply AI platform created by Google, and Unity, a preferred online game engine.
Movement-tracking often requires costly cameras, high-end computer systems, and somebody to put on a skin-tight bodysuit with these foolish plastic balls Velcroed throughout. With this challenge although, all you want is that webcam you’ve had buried in your junk drawer for half a decade:
Utilizing AI for motion-tracking isn’t precisely new; Microsoft’s Kinect was an early instance of a shopper product that utilized machine studying. Nonetheless, it required a number of sensors and price $150 – to not point out it’s lifeless now.
Fleischer and Ayalon’s AI is experimental, however it’s nonetheless thrilling to see real-time movement monitoring occur in any respect with low-cost hardware, particularly contemplating it outputs to a video-game engine already fashionable with AR and VR builders.
A software program answer for motion-tracking primarily based on open-source AI and dirt-cheap webcams may revolutionize a myriad of industries, not simply gaming. It may allow medical professionals analyze motion and gait with out the necessity for extremely specialised hardware, which may enhance the examine of neurological problems and assist with orthopedic rehabilitation.
And, if you happen to watched the video all the best way to the top, you’ll see it’s additionally a cool strategy to visualize dance strikes in actual time. It’s all the time a celebration whenever you carry TensorFlow and a webcam.
The post Programmers use TensorFlow AI to turn any webcam into Microsoft Kinect appeared first on Proinertech.