MIT Artificial Intelligence manages to “see” people through walls

Tool could range from helping older people to living more independently to create applications to monitor diseases such as Parkinson’s and multiple sclerosis.

MIT has created a system similar to X-ray vision, allowing its Artificial Intelligence (AI) platform to track a person through walls, in addition to identifying a specific person from a group of 100 people  per means of wireless. The area responsible for the tool is the Computer Science Laboratory and Artificial Intelligence of the university, which named the RF-Pose technology.

According to MIT News, RF-Pose can be used to monitor diseases such as Parkinson’s, multiple sclerosis, and muscular dystrophy, improving understanding of disease progression and helping physicians adjust their medications correctly. It could also help older people to live more independently, providing additional safety for monitoring falls, injuries, and changes in activity patterns. The team is currently working with physicians to explore health applications.

The research “Estimation of Human Positioning Through the Wall Using Radio Signals” calls the RF-Pose “a solution that harnesses radio signals to accurately track the human 2D pose through walls and obstructions.”

“In addition to detecting the movement, the authors also showed that they could use wireless signals to accurately identify 83% of someone in a 100 person queue. This capability can be particularly useful for the application of search and rescue operations, as well as being useful for knowing the identity of specific people, “says MIT News.

RF-Pose can also be used “for new classes of video games where players move around the house or on search and rescue missions to help locate survivors.”

The system relies on the fact that the signal is not interrupted by walls. The team used the AI ​​”to teach wireless devices to notice the postures and movement of people, even on the other side of the wall. Researchers use a neural network to analyze the radio signals they reflect on people’s bodies and then create a dynamic puppet that walks, sits, and moves its limbs as one performs the same actions. ” Check the video bellow:

Future plans for RF-Pose

The researchers behind RF-Pose are currently working on 3D technology. with it, it would be possible to understand even smaller movements, such as seeing “if the hands of an elderly person are shaking enough so that they can check-up. 😲 ”

In the future, they plan to develop a “consent mechanism” as a countermeasure to block surveillance. This would involve a person doing “a specific set of movements to monitor the environment”.


Leave a Reply





Spanish creates autonomous printer to write in the sand

Scientists create first truly functional invisibility device