We all know very well that for a long time, X-ray vision has been a
resource in fantasy and science fiction novels. But recently, a team of
researchers has developed a new AI that can easily see through walls and
track people’s movement.
For a long time, X-ray vision has been a resource in fantasy and science fiction novels. But it was always there. As recently, a team led by Dina Katabi has taken an important step in the evolution of this technology.
His latest project, RF-Pose, uses artificial intelligence (AI) to teach wireless devices to detect the postures and movements of people, even from the other side of the wall. To do this, the Katabi’s team has designed a neural network that analyzes the radio signals that bounce off people’s bodies. With this information, they create a dynamic figure that walks, stops, sits and moves their limbs as the person performs those actions.
The team says the system could be used to study diseases such as Parkinson’s and multiple sclerosis (MS), providing a better understanding of the progression of the disease and allowing doctors to adjust medications accordingly.
Since the cameras cannot see through the walls, the network was never explicitly trained in data linked to what happens on the other side of the wall, which made it particularly surprising for the MIT team, is that the network he could generalize his knowledge to be able to handle it.
In addition to detecting movement, the authors also demonstrated that they could use wireless signals to accurately identify someone 83% of the time among a group of 100 people. This capability could be particularly useful for the application of search and rescue operations, when it may be useful to know the identity of specific people.
The preview, entitled Through-Wall Human Pose Estimation Using Radio Signals, will be presented at the Conference on Computer Vision and Pattern Recognition.
So, what do you think about this? Simply share all your views and thoughts in the comment section below.
This New AI Tech Can See Through Walls And Track People’s Movement
For a long time, X-ray vision has been a resource in fantasy and science fiction novels. But it was always there. As recently, a team led by Dina Katabi has taken an important step in the evolution of this technology.
His latest project, RF-Pose, uses artificial intelligence (AI) to teach wireless devices to detect the postures and movements of people, even from the other side of the wall. To do this, the Katabi’s team has designed a neural network that analyzes the radio signals that bounce off people’s bodies. With this information, they create a dynamic figure that walks, stops, sits and moves their limbs as the person performs those actions.
The team says the system could be used to study diseases such as Parkinson’s and multiple sclerosis (MS), providing a better understanding of the progression of the disease and allowing doctors to adjust medications accordingly.
It
could also help older people to live more independently while providing
the added security of being able to detect falls, injuries and changes
in activity patterns.
“Just as mobile phones and Wi-Fi routers
have become essential parts of today’s homes, I believe that wireless
technologies like these will help propel the houses of the future,”
Katabi said in a statement.Since the cameras cannot see through the walls, the network was never explicitly trained in data linked to what happens on the other side of the wall, which made it particularly surprising for the MIT team, is that the network he could generalize his knowledge to be able to handle it.
In addition to detecting movement, the authors also demonstrated that they could use wireless signals to accurately identify someone 83% of the time among a group of 100 people. This capability could be particularly useful for the application of search and rescue operations, when it may be useful to know the identity of specific people.
The preview, entitled Through-Wall Human Pose Estimation Using Radio Signals, will be presented at the Conference on Computer Vision and Pattern Recognition.
So, what do you think about this? Simply share all your views and thoughts in the comment section below.
No comments:
Post a Comment