Meta Unveils Wristband That Controls Computers With Muscle Signals (meta.com) 14
Meta researchers published findings in Nature Wednesday detailing a wristband prototype that controls computers through hand gestures by reading electrical signals from forearm muscles. The device uses surface electromyography to detect signals from alpha motor neurons in the spinal cord that connect to muscle fibers, allowing users to move cursors with wrist turns, open applications with thumb-to-forefinger taps, and write text by tracing letters in the air.
The technology, developed at Meta's Reality Labs, trained neural networks on data from 10,000 participants to identify common muscle signal patterns. The wristband works without individual calibration across most users and can detect intended movements before physical motion occurs. Meta demonstrated the device controlling its Orion augmented reality glasses last fall and plans product integration over the next few years.
The technology, developed at Meta's Reality Labs, trained neural networks on data from 10,000 participants to identify common muscle signal patterns. The wristband works without individual calibration across most users and can detect intended movements before physical motion occurs. Meta demonstrated the device controlling its Orion augmented reality glasses last fall and plans product integration over the next few years.
Ok... (Score:2)
Re: (Score:2)
Yep, I tried one about that long ago. Like most of their tech they didn’t invent it but bought it from someone else.
Re: (Score:2)
Did it work reliably for regular daily use by everyday people?
Was its form factor acceptable for regular daily use by everyday people?
Re: (Score:1)
I used a "Myo Armband" briefly, 10 years ago. It combined the sEMG sensing with accelerometers, and mostly worked. The gesture recognition was pretty poor, but I don't know if that was due to limitations in sensing, or in algorithms. You could use it for cueing slides in a presentation, with maybe a 95% success rate.
Re: (Score:2)
The signals at the wrist are pretty shit and determining intermediate finger poses is tricky. You really do need some trained signal processing.
Re: (Score:1)
The "Myo Armband", by Thalmic Labs: https://www.cnet.com/reviews/m... [cnet.com]
Their patents/IP appear to have been passed to "CTRL Labs", and then to Meta/Oculus, hence why Meta is now making the announcements.
Honestly, I think that this is a great use case for neural networks and machine learning. It could solve complex DSP problems without writing massive amounts of algorithms, and collecting training data doesn't involve industrial-scale theft of IP.
Is there a porno version? (Score:3)
however on the down side (Score:2)
Re: (Score:2)
You can also access every user's complete dataset from Meta's Customer Biometrics, Location And Masturbatory Habits DB.
Unfortunately that is only accessible for Meta employees above the rank of Junior Sanitory Cleaner, IIA.
Or if you have the password, which is lk234d24jh77fkaj
Re: (Score:1)
I've got your controller right here (Score:2)
You KNOW what people are going to strap this to....
Meanwhile... (Score:2)
Apple Vision Pro (yes I know it's expensive, it will get cheaper) reads hand gestures with wearing anything other than the headset.
Reminds me of when Jobs dunked on stylus's saying the iPhone touch screen used "the stylus everyone was born with and always has with them."