Research Students from MIT Computer Science and Artificial Intelligence Laboratory (CSAIL) have managed to create a system named RF-Pose which enables seeing through walls by using an RF transmitter and receiving antennas. Eventhough this technique is not entirely novel, the Neural Network used to better decipher what’s received on those antennas is a new approach. The neural networks helps to provide pose estimations of where people’s heads, shoulders, elbows, and other body parts are, and a with a little bit of processing turns that into skeletal figures.
Following video demonstrates the system in action.
If you are curious about the nitty gritties of the system you can the read the research paper named Through-Wall Human Pose Estimation Using Radio Signals.