Ever wish you had superpowers that allows you look through solid objects? Turns out that we don’t really need superpowers to do that. The folks at MIT have recently come up with a way that can sort of let you do that, and that is by harnessing the power of WiFi which in turn allows the computer to identify the person behind a solid object like a wall.
How does it work? The concept is actually pretty simple. Basically the device will send out WiFi signals from one side of the wall. The signals will then hit the person behind the wall and those reflections will be captured by the device, and then sent to a software to be cleaned up in which a rough shape of the person’s body can be seen.
Now this only gives us a clue that a person is behind the wall, but how does it identify who it is? Speaking to Gizmodo, one of the researchers by the name of Fadel Adi explains that by using machine language, the algorithms can spot subtle differences in body shapes, which in turn will allow it to identify one person from another.
According to Adi, “[W]e use the captured human silhouettes from our reconstruction algorithm [to] train a classifier on these silhouettes which allows us to distinguish between people. The classifier captures features like height and body builds, which allows us to distinguish between people using RF-Capture.”
As far as its uses are concerned, the team envisions how it could be used as a nursing tool for the elderly. For example it could be placed in homes of the elderly and will keep scanning for them, and should it detect that they have fallen over, it could even be programmed to call 911 on their behalf.