A system that tracks human poses in 3D, the perfect spy

I recently told you that Google was launching an Artificial Intelligence system Thought of teaching to dance and synchronize movements, today I will talk about something similar that, combined with other technologies, can determine what future movement tracking will be like.

Currently, to detect movements wirelessly, WiFi signals, Lidar and similar options are used. A signal is emitted, you wait for it to bounce, and by calculating the time it takes to go and return, you know the distance you are at. If this distance is modified at each point, it is known that the object is moving and it can be calculated how it is moving.

These techniques can be used for surveillance, to track sleep, or to monitor patients or athletes, as well as to create more realistic gaming experiences.

Now it has been researchers from Florida State University, Trinity University and Rutgers University who have presented Winect, a new wireless sensing system that can track the poses of humans in 3D while performing a wide range of physical activities freely.

They published it in arXiv and they will present it in more detail at the ACM Conference on Interactive Technologies.

Until now the systems were based on models that were previously trained with known activities, so that if someone moved doing something that was not registered, the system could not know what they were doing. What they have presented now allows tracking of free movements, not tracked before.

They also use Wi-Fi signals to detect activity, and can apply it to virtual reality (VR) applications, augmented fitness, and video game development.

What can Winect do

This new system can examine movements in home environments using WiFi, as I mentioned before, and can do so using existing devices, such as computers, smart TVs or speakers. These devices emit the signal and deep learning techniques are used to create a digital version of the 3D full-body movements of the people in the room.

Independent movement of the head, spine, shoulders, elbows, wrists, hips, knees, and ankles can be tracked.

It first analyzes the environment, and then it analyzes the signals reflected in the human body of the person.

Winect thus creates a digital version of the poses of the body of a human user in 3D, a 3D digital skeleton imitating the movements made by the person.

Being WiFi, it can also detect through walls, as it does not need direct vision, like Kinect-type systems, which use a camera.

In the future, systems could be created that, in addition to tracking human activities, can predict intentions, something extremely important in the world of surveillance and security.

Source link

Lenny Li

I started to play with tech since middle school. Smart phones, laptops and gadgets are all about my life. Besides, I am also a big fan of Star War. May the force be with you!

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top button