WiFi devices as perception sensors are invariant to illumination and privacy-friendly comparing to cameras, while are cheaper, smaller, and more power efficient than radars and LiDARs. In this project, we present the first work that given 1D data received at WiFi antennas, it is possible to reconstruct 2D fine-grained spatial information of human bodies. Our Person-in-WiFi approach is based on off-the-shell WiFi antennas lined-up as regular house-hold WiFi routers, making it very easy to develop perception applications in any indoor environments such as warehouse, hospital, office and home.

Related papers:

Person-in-WiFi: Fine-grained Person Perception using WiFi
F Wang, S Zhou, S Panev, J Han, D Huang
arXiv preprint arXiv:1904.00276

CSI-Net: Unified Human Body Characterization and Action Recognition
F Wang, J Han, S Zhang, X He, D Huang
arXiv preprint arXiv:1810.03064

Can WiFi Estimate Person Pose?
F Wang, S Panev, Z Dai, J Han, D Huang
arXiv preprint arXiv:1904.00277

WiPIN: Operation-free Person Identification using WiFi Signals
F Wang, J Han, Z Dai, H Ding, D Huang
arXiv preprint arXiv:1810.04106