RF-Pose (software for viewing through walls)

2018: Creation of software

In June, 2018 specialists of laboratory of computer sciences and artificial intelligence of the Massachusetts Institute of Technology (MIT Computer Science and Artificial Intelligence Laboratory, CSAIL) announced creation of an AI system capable "see" movements and distinguish poses of people through walls.

Within the project under the name RF-Pose scientists used a neuronet for the analysis of the radio signals reflected from human bodies. On the basis of these data an AI system learned to understand how people move and to create the dynamic model reproducing gestures, poses and movements of the person even when that is behind a wall.

Artificial intelligence taught to see people through walls

Specialists say that the RF-Pose technology can be useful to patients, the suffering such diseases as Parkinson's disease, multiple sclerosis and muscular dystrophy. Using technology doctors will be able to understand better course of disease and to adjust drug treatment.

Besides, RF-Pose can be used for delicate monitoring of people of old age and prevention of their falling and injuries. According to the press service of MIT, the team of scientists of institute together with doctors already develops scenarios of use of technology in health care.

The RF-Pose project manager professor Dina Katabi called important advantage of technology that patients should not put on any sensors and devices needing regular recharge.

In addition to medical application, RF-Pose can form the basis of new type of computer games. Besides, during the experiments it was revealed that the technology allows to distinguish with an accuracy of 83% people on the nature of their gait and movements. Believe in MIT that this function can it is useful during search and rescue transactions for identification of people.^[1]

Robotics

Notes