For a while now researchers at MIT and several other universities have been investigating methods for using frequencies in the WiFi bands to see through walls using a form of low power radar. The basic concept is to track and process the reflections of these signals from peoples bodies.
Recently researchers at MIT have taken this idea a step further, combining the radar results with machine learning in a project they call RF-Pose. The result is the ability to recreate and track full human post information through walls. The abstract from their paper reads:
This paper demonstrates accurate human pose estimation through walls and occlusions. We leverage the fact that wireless signals in the WiFi frequencies traverse walls and reflect off the human body. We introduce a deep neural network approach that parses such radio signals to estimate 2D poses. Since humans cannot annotate radio signals, we use state-of-the-art vision model to provide cross-modal supervision.
Specifically, during training the system uses synchronized wireless and visual inputs, extracts pose information from the visual stream, and uses it to guide the training process. Once trained, the network uses only the wireless signal for pose estimation. We show that, when tested on visible scenes, the radio-based system is almost as accurate as the vision-based system used to train it. Yet, unlike vision-based pose estimation, the radio-based system can estimate 2D poses through walls despite never trained on such scenarios.
The hope is that this technology could one day be used as a replacement for camera based computer vision. It would be a non-intrusive method for applications like gaming, monitoring the elderly for falls, motion capture during film making without the need for suits and of course for gathering data on peoples movements.
It is not mentioned in the paper, but it is likely that they are using some sort of SDR like a USRP for receiving the signals. It's possible that a lower resolution system could be set up cheaply with a HackRF and some passive radar software.