Before we know it, X-ray vision may be more than a comic book fantasy. We’re still a far way off from the fictional superpower our inner nerds so deeply desire, but researchers at MIT’s Computer Science and Artificial Intelligence Lab (CSAIL) are getting pretty damn close.
A team at the university just shared its paper on how they’ve developed the technology to not only detect human movements through walls – a development made back in 2013 – but also to see individuals’ silhouettes, distinguish different people from each other and even sense different postures.
“We’re working to turn this technology into an in-home device that can call 911 if it detects that a family member has fallen unconscious.”
The group of researchers includes MIT PhD student and lead author Fadel Adib, Professor and Director of Wireless@MIT Dina Katabi, Professor Frédo Durand, PhD student Chen-Yu Hsu and intern Hongzi Mao. They’ve created a device that uses wireless signals to see and track people’s silhouettes with shocking accuracy.
Their technology emits these signals, which go through walls, bounce off of human forms on the other side and give feedback to the device. To piece together more complete silhouettes, the device takes in these wireless reflections as the human forms move. This allows the technology to make out different body parts, which are then pieced together using sophisticated algorithms.
What could it do?
Their research paves the way for a variety of progressive uses. The team has thrown out the possibility of applying their technology to an extremely wide range of things from gaming to smart home control.
For instance, because their creation is able to track precise bodily movements, they believe it’d be perfect for Hollywood. Filmmakers would no longer have to hook actors up with sensors all over their bodies, as the device would be able to do all of the work wirelessly.
“Today actors have to wear markers on their bodies and move in a specific room full of cameras,” explained Adib. “RF Capture would enable motion capture without body sensors and could track actors’ movements even if they are behind furniture or walls.”
At the same time, because the technology is capable of distinguishing people’s positioning – so standing versus sitting or lying down, for example – the team maintains it could be key in helping with individuals’ personal safety within their homes.
“We’re working to turn this technology into an in-home device that can call 911 if it detects that a family member has fallen unconscious,” said Katabi.
This application is especially important in elder care, and that’s exactly what the team is trying to tackle now. They’ve seen great potential in this particular use and have been working on developing a commercial product called Emerald. The device they’re creating is supposed to be much more than simply detecting when an elderly person has fallen; the researchers are planning on also having the technology sense when someone is about to fall and prevent it.