Short Bytes: MIT researchers have developed an RF capture device that could identify human silhouettes through the wall by using WiFI signals. The device is machine trained to recognize humans while profiling the major body parts.
The research team of the institute’s Computer Science and Artificial Intelligence Lab had been using wireless signals for quite some time. Now they have come up with the RF-Capture technique that is trained to recognize different people based on their WiFi imagery.
The technology is intricate, but the idea is fairly simple though. The WiFi signal coming through the wall will be reflected by a human body and then captured by the device. The captured signals are then processed and the noise removed through a software.
As the data is received, it is fed to the algorithms that have been trained through machine learning techniques to detect the bodily features and release a coarse human model with major body parts as head chest arms and feet along with classification of different silhouettes on basis of height and other subtleties.
MIT, in spite of being the most insecure university campus in America, is one of the main centers of innovation and developments of the world. Like with this one, the researchers are aiming to provide help to the elderly. Through this device, people can monitor their grandparents in the other room and even call 911 in case of emergency.
The technology is still in developing stage and researchers are trying to minimize the privacy concerns that could arise with this device.
Watch the demo of RF capture by MIT: