Motion tracking technology is no longer surprising, but a team of engineers from the Massachusetts Institute of Technology (MIT) was able to create a device that can track a person through walls. Moreover, the device turned out to be not bulky at all.
The device was named RF-Pose. To teach the algorithm new tricks, MIT employees tracked people's movements using radar and video cameras. They recorded walking, conversation between people, sitting, standing and waiting positions, as well as opening doors. Then, using a not very complex computer program, the image was transformed into a skeletal model of each recorded situation. These models, together with the radio signal, were studied by the AI, and thus he learned to recognize the relationship between the readings of the radar and what actions a person or group of people was performing at a given moment.
As a result, MIT specialists created an algorithm that can show the movement of people behind a wall or other obstacle in almost real time. It is worth noting that at the moment the system is capable of "producing" only a two-dimensional image, but over time the system can be optimized so that it can monitor in three dimensions.
The creators also plan to teach AI to recognize more complex movements, such as fine motor skills of the hands and fingers, since now it is only possible to observe the movement of the limbs and trunk. The authors themselves argue that their system can find application in a variety of areas. From quite logical observations of crowded places to medical institutions and places of detention.
Vladimir Kuznetsov