New AI technology sees people’s movement through walls

Seeing through walls via x-rays was possible but it wasn’t through wireless radio signals. Researchers have now made this possible by creating a new technology via artificial intelligence to see people and their postures through walls.

Scientists at the Massachusetts Institute of Technology’s Computer Science and Artificial Intelligence Laboratory (MIT CSAIL) have developed a system that sees people through walls. The mechanism recreates people’s poses while they move through wireless radio signals and recreate the body like a plain stick figure called ‘RF-Pose’.

Tech Crunch explained the team used a neural network to analyze radio signals that bounces off people’s bodies and creates a dynamic stick figure that can walk, sit, stop and move its limbs as the person performs those actions.

Researchers create world’s first psychopathic AI using gruesome images

The neural network was trained by showing machine, a video of a person walking next to the RF interference they made as they moved. Then, they overlaid stick figures on the movement and trained the network to do the same automatically. Since RF signals are everywhere, it was easier to use than other technologies. Also, the team never trained the system to see through walls, but it was able to ‘generalize its knowledge to be able to handle through-wall movement.’

The RF-Pose made identification with a great degree of accuracy, regardless of the fact that no cameras were used for making comparisons, wrote New Atlas.

Video Courtesy: MITCSAIL

This new AI technology is less harmful way to examine people for health and safety purposes.  As per the team, this mechanism can be used for monitoring people suffering from Parkinson’s disease or multiple sclerosis (MS), which will help doctors know more about the disease progression, enabling them to prescribe medications accordingly. It can also keep an eye on elderly patients and give them a more independent life.

Researchers also clarified that the data gathered of the subjects was after their consent and is also completely encrypted to ensure privacy. The primary focus of the team is to make use of this technology for healthcare, allowing passive monitoring of an individual person without the need of cameras or other intrusions.

“For future real-world applications, the team plans to implement a ‘consent mechanism’ in which the person who installs the device is cued to do a specific set of movements in order for it to begin to monitor the environment,” the researchers wrote.